diff --git a/docs/_posts/Cabir40/2023-01-30-t5_1zha5ono_en.md b/docs/_posts/Cabir40/2023-01-30-t5_1zha5ono_en.md new file mode 100644 index 00000000000000..cb0df9b3906c3f --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_1zha5ono_en.md @@ -0,0 +1,91 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from tscholak) +author: John Snow Labs +name: t5_1zha5ono +date: 2023-01-30 +tags: [en, open_source, t5] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `1zha5ono` is a English model originally trained by `tscholak`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_1zha5ono_en_4.3.0_3.0_1675095935006.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_1zha5ono_en_4.3.0_3.0_1675095935006.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols(["text"]) \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_1zha5ono","en") \ + .setInputCols(["document"]) \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_1zha5ono","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_1zha5ono| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|947.0 MB| + +## References + +- https://huggingface.co/tscholak/1zha5ono +- https://arxiv.org/abs/2109.05093 +- https://github.com/google-research/text-to-text-transfer-transformer/blob/main/released_checkpoints.md#lm-adapted-t511lm100k +- https://yale-lily.github.io/spider +- https://github.com/google-research/text-to-text-transfer-transformer/blob/main/released_checkpoints.md#lm-adapted-t511lm100k +- https://github.com/ElementAI/picard +- https://github.com/ElementAI/picard +- https://arxiv.org/abs/2109.05093 +- https://github.com/ElementAI/picard diff --git a/docs/_posts/Cabir40/2023-01-30-t5_apoorva_k2t_test_en.md b/docs/_posts/Cabir40/2023-01-30-t5_apoorva_k2t_test_en.md new file mode 100644 index 00000000000000..e4962c24423a9c --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_apoorva_k2t_test_en.md @@ -0,0 +1,84 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from Apoorva) +author: John Snow Labs +name: t5_apoorva_k2t_test +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `k2t-test` is a English model originally trained by `Apoorva`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_apoorva_k2t_test_en_4.3.0_3.0_1675103912014.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_apoorva_k2t_test_en_4.3.0_3.0_1675103912014.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_apoorva_k2t_test","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_apoorva_k2t_test","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_apoorva_k2t_test| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|275.8 MB| + +## References + +- https://huggingface.co/Apoorva/k2t-test \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_arabic_text_summarization_ar.md b/docs/_posts/Cabir40/2023-01-30-t5_arabic_text_summarization_ar.md new file mode 100644 index 00000000000000..b46d42cabde208 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_arabic_text_summarization_ar.md @@ -0,0 +1,84 @@ +--- +layout: model +title: Arabic T5ForConditionalGeneration Cased model (from malmarjeh) +author: John Snow Labs +name: t5_arabic_text_summarization +date: 2023-01-30 +tags: [ar, open_source, t5, tensorflow] +task: Text Generation +language: ar +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-arabic-text-summarization` is a Arabic model originally trained by `malmarjeh`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_arabic_text_summarization_ar_4.3.0_3.0_1675107890347.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_arabic_text_summarization_ar_4.3.0_3.0_1675107890347.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_arabic_text_summarization","ar") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_arabic_text_summarization","ar") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_arabic_text_summarization| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|ar| +|Size:|1.4 GB| + +## References + +- https://huggingface.co/malmarjeh/t5-arabic-text-summarization \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_arat5_base_title_generation_ar.md b/docs/_posts/Cabir40/2023-01-30-t5_arat5_base_title_generation_ar.md new file mode 100644 index 00000000000000..6cb5612d1626e1 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_arat5_base_title_generation_ar.md @@ -0,0 +1,86 @@ +--- +layout: model +title: Arabic T5ForConditionalGeneration Base Cased model (from UBC-NLP) +author: John Snow Labs +name: t5_arat5_base_title_generation +date: 2023-01-30 +tags: [ar, open_source, t5] +task: Text Generation +language: ar +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `AraT5-base-title-generation` is a Arabic model originally trained by `UBC-NLP`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_arat5_base_title_generation_ar_4.3.0_3.0_1675096301207.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_arat5_base_title_generation_ar_4.3.0_3.0_1675096301207.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols(["text"]) \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_arat5_base_title_generation","ar") \ + .setInputCols(["document"]) \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_arat5_base_title_generation","ar") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_arat5_base_title_generation| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|ar| +|Size:|1.4 GB| + +## References + +- https://huggingface.co/UBC-NLP/AraT5-base-title-generation +- https://aclanthology.org/2022.acl-long.47/ +- https://doi.org/10.14288/SOCKEYE +- https://www.tensorflow.org/tfrc diff --git a/docs/_posts/Cabir40/2023-01-30-t5_autotrain_amazon_text_sum_730222226_en.md b/docs/_posts/Cabir40/2023-01-30-t5_autotrain_amazon_text_sum_730222226_en.md new file mode 100644 index 00000000000000..31cb768085d399 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_autotrain_amazon_text_sum_730222226_en.md @@ -0,0 +1,84 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from yogi) +author: John Snow Labs +name: t5_autotrain_amazon_text_sum_730222226 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `autotrain-amazon_text_sum-730222226` is a English model originally trained by `yogi`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_autotrain_amazon_text_sum_730222226_en_4.3.0_3.0_1675099816649.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_autotrain_amazon_text_sum_730222226_en_4.3.0_3.0_1675099816649.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_autotrain_amazon_text_sum_730222226","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_autotrain_amazon_text_sum_730222226","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_autotrain_amazon_text_sum_730222226| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|284.9 MB| + +## References + +- https://huggingface.co/yogi/autotrain-amazon_text_sum-730222226 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_autotrain_inference_probability_3_900329401_en.md b/docs/_posts/Cabir40/2023-01-30-t5_autotrain_inference_probability_3_900329401_en.md new file mode 100644 index 00000000000000..3cc9993fc31499 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_autotrain_inference_probability_3_900329401_en.md @@ -0,0 +1,84 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from jeremyccollinsmpi) +author: John Snow Labs +name: t5_autotrain_inference_probability_3_900329401 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `autotrain-inference_probability_3-900329401` is a English model originally trained by `jeremyccollinsmpi`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_autotrain_inference_probability_3_900329401_en_4.3.0_3.0_1675099900080.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_autotrain_inference_probability_3_900329401_en_4.3.0_3.0_1675099900080.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_autotrain_inference_probability_3_900329401","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_autotrain_inference_probability_3_900329401","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_autotrain_inference_probability_3_900329401| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|846.1 MB| + +## References + +- https://huggingface.co/jeremyccollinsmpi/autotrain-inference_probability_3-900329401 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_autotrain_ms_2_1174443640_en.md b/docs/_posts/Cabir40/2023-01-30-t5_autotrain_ms_2_1174443640_en.md new file mode 100644 index 00000000000000..bbbc802e7011cf --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_autotrain_ms_2_1174443640_en.md @@ -0,0 +1,84 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from benjamyu) +author: John Snow Labs +name: t5_autotrain_ms_2_1174443640 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `autotrain-ms-2-1174443640` is a English model originally trained by `benjamyu`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_autotrain_ms_2_1174443640_en_4.3.0_3.0_1675099983295.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_autotrain_ms_2_1174443640_en_4.3.0_3.0_1675099983295.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_autotrain_ms_2_1174443640","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_autotrain_ms_2_1174443640","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_autotrain_ms_2_1174443640| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|916.0 MB| + +## References + +- https://huggingface.co/benjamyu/autotrain-ms-2-1174443640 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_banglat5_banglaparaphrase_bn.md b/docs/_posts/Cabir40/2023-01-30-t5_banglat5_banglaparaphrase_bn.md new file mode 100644 index 00000000000000..7fd391d596ecea --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_banglat5_banglaparaphrase_bn.md @@ -0,0 +1,86 @@ +--- +layout: model +title: Bangla T5ForConditionalGeneration Cased model (from csebuetnlp) +author: John Snow Labs +name: t5_banglat5_banglaparaphrase +date: 2023-01-30 +tags: [bn, open_source, t5, tensorflow] +task: Text Generation +language: bn +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `banglat5_banglaparaphrase` is a Bangla model originally trained by `csebuetnlp`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_banglat5_banglaparaphrase_bn_4.3.0_3.0_1675100065577.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_banglat5_banglaparaphrase_bn_4.3.0_3.0_1675100065577.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_banglat5_banglaparaphrase","bn") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_banglat5_banglaparaphrase","bn") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_banglat5_banglaparaphrase| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|bn| +|Size:|1.0 GB| + +## References + +- https://huggingface.co/csebuetnlp/banglat5_banglaparaphrase +- https://github.com/csebuetnlp/BanglaNLG +- https://github.com/csebuetnlp/normalizer \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_banglat5_nmt_bn2en_xx.md b/docs/_posts/Cabir40/2023-01-30-t5_banglat5_nmt_bn2en_xx.md new file mode 100644 index 00000000000000..e242703498fb5a --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_banglat5_nmt_bn2en_xx.md @@ -0,0 +1,85 @@ +--- +layout: model +title: Multilingual T5ForConditionalGeneration Cased model (from csebuetnlp) +author: John Snow Labs +name: t5_banglat5_nmt_bn2en +date: 2023-01-30 +tags: [bn, en, open_source, t5, xx, tensorflow] +task: Text Generation +language: xx +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `banglat5_nmt_bn_en` is a Multilingual model originally trained by `csebuetnlp`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_banglat5_nmt_bn2en_xx_4.3.0_3.0_1675100164821.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_banglat5_nmt_bn2en_xx_4.3.0_3.0_1675100164821.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_banglat5_nmt_bn2en","xx") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_banglat5_nmt_bn2en","xx") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_banglat5_nmt_bn2en| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|xx| +|Size:|1.0 GB| + +## References + +- https://huggingface.co/csebuetnlp/banglat5_nmt_bn_en +- https://github.com/csebuetnlp/normalizer \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_banglat5_nmt_en2bn_xx.md b/docs/_posts/Cabir40/2023-01-30-t5_banglat5_nmt_en2bn_xx.md new file mode 100644 index 00000000000000..2fef4f57e10475 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_banglat5_nmt_en2bn_xx.md @@ -0,0 +1,85 @@ +--- +layout: model +title: Multilingual T5ForConditionalGeneration Cased model (from csebuetnlp) +author: John Snow Labs +name: t5_banglat5_nmt_en2bn +date: 2023-01-30 +tags: [bn, en, open_source, t5, xx, tensorflow] +task: Text Generation +language: xx +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `banglat5_nmt_en_bn` is a Multilingual model originally trained by `csebuetnlp`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_banglat5_nmt_en2bn_xx_4.3.0_3.0_1675100250892.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_banglat5_nmt_en2bn_xx_4.3.0_3.0_1675100250892.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_banglat5_nmt_en2bn","xx") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_banglat5_nmt_en2bn","xx") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_banglat5_nmt_en2bn| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|xx| +|Size:|1.0 GB| + +## References + +- https://huggingface.co/csebuetnlp/banglat5_nmt_en_bn +- https://github.com/csebuetnlp/normalizer \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_base_amazonreviews_en.md b/docs/_posts/Cabir40/2023-01-30-t5_base_amazonreviews_en.md new file mode 100644 index 00000000000000..42b85693d4253b --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_base_amazonreviews_en.md @@ -0,0 +1,84 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from sumedh) +author: John Snow Labs +name: t5_base_amazonreviews +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-base-amazonreviews` is a English model originally trained by `sumedh`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_base_amazonreviews_en_4.3.0_3.0_1675107991189.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_base_amazonreviews_en_4.3.0_3.0_1675107991189.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_base_amazonreviews","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_base_amazonreviews","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_base_amazonreviews| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|921.3 MB| + +## References + +- https://huggingface.co/sumedh/t5-base-amazonreviews \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_base_askscience_en.md b/docs/_posts/Cabir40/2023-01-30-t5_base_askscience_en.md new file mode 100644 index 00000000000000..9592ea70f67037 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_base_askscience_en.md @@ -0,0 +1,84 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from pszemraj) +author: John Snow Labs +name: t5_base_askscience +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-base-askscience` is a English model originally trained by `pszemraj`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_base_askscience_en_4.3.0_3.0_1675108079371.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_base_askscience_en_4.3.0_3.0_1675108079371.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_base_askscience","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_base_askscience","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_base_askscience| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|1.0 GB| + +## References + +- https://huggingface.co/pszemraj/t5-base-askscience \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_base_bahasa_cased_ms.md b/docs/_posts/Cabir40/2023-01-30-t5_base_bahasa_cased_ms.md new file mode 100644 index 00000000000000..00dfdf28851306 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_base_bahasa_cased_ms.md @@ -0,0 +1,87 @@ +--- +layout: model +title: Malay T5ForConditionalGeneration Base Cased model (from mesolitica) +author: John Snow Labs +name: t5_base_bahasa_cased +date: 2023-01-30 +tags: [ms, open_source, t5, tensorflow] +task: Text Generation +language: ms +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-base-bahasa-cased` is a Malay model originally trained by `mesolitica`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_base_bahasa_cased_ms_4.3.0_3.0_1675108290125.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_base_bahasa_cased_ms_4.3.0_3.0_1675108290125.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_base_bahasa_cased","ms") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_base_bahasa_cased","ms") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_base_bahasa_cased| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|ms| +|Size:|473.3 MB| + +## References + +- https://huggingface.co/mesolitica/t5-base-bahasa-cased +- https://github.com/huseinzol05/malaya/tree/master/pretrained-model/t5/prepare +- https://github.com/google-research/text-to-text-transfer-transformer +- https://github.com/huseinzol05/Malaya/tree/master/pretrained-model/t5 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_base_conversation_en.md b/docs/_posts/Cabir40/2023-01-30-t5_base_conversation_en.md new file mode 100644 index 00000000000000..8e541d0d07566c --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_base_conversation_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from Supiri) +author: John Snow Labs +name: t5_base_conversation +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-base-conversation` is a English model originally trained by `Supiri`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_base_conversation_en_4.3.0_3.0_1675108355643.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_base_conversation_en_4.3.0_3.0_1675108355643.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_base_conversation","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_base_conversation","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_base_conversation| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|926.0 MB| + +## References + +- https://huggingface.co/Supiri/t5-base-conversation +- https://docs.unrealengine.com/5.0/en-US/RenderingFeatures/Nanite/ +- https://www.youtube.com/watch?v=WU0gvPcc3jQ +- https://www.youtube.com/watch?v=Z1OtYGzUoSo +- https://www.personality-database.com/profile/2790/hinata-hyga-naruto-shippden-mbti-personality-type \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_base_define_en.md b/docs/_posts/Cabir40/2023-01-30-t5_base_define_en.md new file mode 100644 index 00000000000000..f60b5e258580e7 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_base_define_en.md @@ -0,0 +1,85 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from marksverdhei) +author: John Snow Labs +name: t5_base_define +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-base-define` is a English model originally trained by `marksverdhei`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_base_define_en_4.3.0_3.0_1675108436852.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_base_define_en_4.3.0_3.0_1675108436852.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_base_define","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_base_define","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_base_define| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|926.2 MB| + +## References + +- https://huggingface.co/marksverdhei/t5-base-define +- https://gist.github.com/marksverdhei/0a13f67e65460b71c05fcf558a6a91ae \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_base_english_japanese_xx.md b/docs/_posts/Cabir40/2023-01-30-t5_base_english_japanese_xx.md new file mode 100644 index 00000000000000..c98b47a7b2e300 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_base_english_japanese_xx.md @@ -0,0 +1,92 @@ +--- +layout: model +title: Multilingual T5ForConditionalGeneration Base Cased model (from sonoisa) +author: John Snow Labs +name: t5_base_english_japanese +date: 2023-01-30 +tags: [en, ja, multilingual, open_source, t5, xx, tensorflow] +task: Text Generation +language: xx +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-base-english-japanese` is a Multilingual model originally trained by `sonoisa`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_base_english_japanese_xx_4.3.0_3.0_1675108659585.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_base_english_japanese_xx_4.3.0_3.0_1675108659585.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_base_english_japanese","xx") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_base_english_japanese","xx") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_base_english_japanese| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|xx| +|Size:|521.2 MB| + +## References + +- https://huggingface.co/sonoisa/t5-base-english-japanese +- https://en.wikipedia.org +- https://ja.wikipedia.org +- https://oscar-corpus.com +- http://data.statmt.org/cc-100/ +- http://data.statmt.org/cc-100/ +- https://github.com/sonoisa/t5-japanese +- https://creativecommons.org/licenses/by-sa/4.0/deed.ja +- http://commoncrawl.org/terms-of-use/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_base_finetuned_bias_en.md b/docs/_posts/Cabir40/2023-01-30-t5_base_finetuned_bias_en.md new file mode 100644 index 00000000000000..1c1da0210d9cef --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_base_finetuned_bias_en.md @@ -0,0 +1,85 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from erickfm) +author: John Snow Labs +name: t5_base_finetuned_bias +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-base-finetuned-bias` is a English model originally trained by `erickfm`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_base_finetuned_bias_en_4.3.0_3.0_1675108728968.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_base_finetuned_bias_en_4.3.0_3.0_1675108728968.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_base_finetuned_bias","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_base_finetuned_bias","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_base_finetuned_bias| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|922.1 MB| + +## References + +- https://huggingface.co/erickfm/t5-base-finetuned-bias +- https://github.com/rpryzant/neutralizing-bias \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_base_finetuned_break_data_en.md b/docs/_posts/Cabir40/2023-01-30-t5_base_finetuned_break_data_en.md new file mode 100644 index 00000000000000..38a68493df2283 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_base_finetuned_break_data_en.md @@ -0,0 +1,91 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from mrm8488) +author: John Snow Labs +name: t5_base_finetuned_break_data +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-base-finetuned-break_data` is a English model originally trained by `mrm8488`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_base_finetuned_break_data_en_4.3.0_3.0_1675108822999.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_base_finetuned_break_data_en_4.3.0_3.0_1675108822999.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_base_finetuned_break_data","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_base_finetuned_break_data","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_base_finetuned_break_data| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|878.4 MB| + +## References + +- https://huggingface.co/mrm8488/t5-base-finetuned-break_data +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/pdf/1910.10683.pdf +- https://i.imgur.com/jVFMMWR.png +- https://colab.research.google.com/github/patil-suraj/exploring-T5/blob/master/T5_on_TPU.ipynb +- https://twitter.com/psuraj28 +- https://twitter.com/mrm8488 +- https://www.linkedin.com/in/manuel-romero-cs/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_base_finetuned_break_data_question_retrieval_en.md b/docs/_posts/Cabir40/2023-01-30-t5_base_finetuned_break_data_question_retrieval_en.md new file mode 100644 index 00000000000000..aa461774c9608a --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_base_finetuned_break_data_question_retrieval_en.md @@ -0,0 +1,91 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from mrm8488) +author: John Snow Labs +name: t5_base_finetuned_break_data_question_retrieval +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-base-finetuned-break_data-question-retrieval` is a English model originally trained by `mrm8488`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_base_finetuned_break_data_question_retrieval_en_4.3.0_3.0_1675108919499.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_base_finetuned_break_data_question_retrieval_en_4.3.0_3.0_1675108919499.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_base_finetuned_break_data_question_retrieval","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_base_finetuned_break_data_question_retrieval","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_base_finetuned_break_data_question_retrieval| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|878.5 MB| + +## References + +- https://huggingface.co/mrm8488/t5-base-finetuned-break_data-question-retrieval +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/pdf/1910.10683.pdf +- https://i.imgur.com/jVFMMWR.png +- https://colab.research.google.com/github/patil-suraj/exploring-T5/blob/master/T5_on_TPU.ipynb +- https://twitter.com/psuraj28 +- https://twitter.com/mrm8488 +- https://www.linkedin.com/in/manuel-romero-cs/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_base_finetuned_cuad_en.md b/docs/_posts/Cabir40/2023-01-30-t5_base_finetuned_cuad_en.md new file mode 100644 index 00000000000000..525be0fb718667 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_base_finetuned_cuad_en.md @@ -0,0 +1,84 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from mrm8488) +author: John Snow Labs +name: t5_base_finetuned_cuad +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `T5-base-finetuned-cuad` is a English model originally trained by `mrm8488`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_base_finetuned_cuad_en_4.3.0_3.0_1675099512929.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_base_finetuned_cuad_en_4.3.0_3.0_1675099512929.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_base_finetuned_cuad","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_base_finetuned_cuad","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_base_finetuned_cuad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|913.5 MB| + +## References + +- https://huggingface.co/mrm8488/T5-base-finetuned-cuad \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_base_finetuned_span_sentiment_extraction_en.md b/docs/_posts/Cabir40/2023-01-30-t5_base_finetuned_span_sentiment_extraction_en.md new file mode 100644 index 00000000000000..2726903392809b --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_base_finetuned_span_sentiment_extraction_en.md @@ -0,0 +1,93 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from mrm8488) +author: John Snow Labs +name: t5_base_finetuned_span_sentiment_extraction +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-base-finetuned-span-sentiment-extraction` is a English model originally trained by `mrm8488`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_base_finetuned_span_sentiment_extraction_en_4.3.0_3.0_1675109003319.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_base_finetuned_span_sentiment_extraction_en_4.3.0_3.0_1675109003319.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_base_finetuned_span_sentiment_extraction","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_base_finetuned_span_sentiment_extraction","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_base_finetuned_span_sentiment_extraction| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|905.9 MB| + +## References + +- https://huggingface.co/mrm8488/t5-base-finetuned-span-sentiment-extraction +- https://twitter.com/AND__SO +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://www.kaggle.com/c/tweet-sentiment-extraction +- https://arxiv.org/pdf/1910.10683.pdf +- https://www.kaggle.com/c/tweet-sentiment-extraction +- https://github.com/enzoampil/t5-intro/blob/master/t5_qa_training_pytorch_span_extraction.ipynb +- https://github.com/enzoampil +- https://twitter.com/mrm8488 +- https://www.linkedin.com/in/manuel-romero-cs/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_base_finetuned_squadv2_en.md b/docs/_posts/Cabir40/2023-01-30-t5_base_finetuned_squadv2_en.md new file mode 100644 index 00000000000000..ceefdf4a7088a4 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_base_finetuned_squadv2_en.md @@ -0,0 +1,91 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from mrm8488) +author: John Snow Labs +name: t5_base_finetuned_squadv2 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-base-finetuned-squadv2` is a English model originally trained by `mrm8488`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_base_finetuned_squadv2_en_4.3.0_3.0_1675109106503.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_base_finetuned_squadv2_en_4.3.0_3.0_1675109106503.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_base_finetuned_squadv2","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_base_finetuned_squadv2","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_base_finetuned_squadv2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|861.2 MB| + +## References + +- https://huggingface.co/mrm8488/t5-base-finetuned-squadv2 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://rajpurkar.github.io/SQuAD-explorer/ +- https://arxiv.org/pdf/1910.10683.pdf +- https://i.imgur.com/jVFMMWR.png +- https://colab.research.google.com/github/patil-suraj/exploring-T5/blob/master/T5_on_TPU.ipynb +- https://twitter.com/mrm8488 +- https://www.linkedin.com/in/manuel-romero-cs/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_base_finetuned_summarize_news_en.md b/docs/_posts/Cabir40/2023-01-30-t5_base_finetuned_summarize_news_en.md new file mode 100644 index 00000000000000..5469b994131425 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_base_finetuned_summarize_news_en.md @@ -0,0 +1,94 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from mrm8488) +author: John Snow Labs +name: t5_base_finetuned_summarize_news +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-base-finetuned-summarize-news` is a English model originally trained by `mrm8488`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_base_finetuned_summarize_news_en_4.3.0_3.0_1675109192613.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_base_finetuned_summarize_news_en_4.3.0_3.0_1675109192613.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_base_finetuned_summarize_news","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_base_finetuned_summarize_news","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_base_finetuned_summarize_news| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|924.2 MB| + +## References + +- https://huggingface.co/mrm8488/t5-base-finetuned-summarize-news +- https://github.com/abhimishra91 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://www.kaggle.com/sunnysai12345/news-summary +- https://arxiv.org/pdf/1910.10683.pdf +- https://i.imgur.com/jVFMMWR.png +- https://www.kaggle.com/sunnysai12345/news-summary +- https://github.com/abhimishra91/transformers-tutorials/blob/master/transformers_summarization_wandb.ipynb +- https://github.com/abhimishra91 +- https://twitter.com/mrm8488 +- https://www.linkedin.com/in/manuel-romero-cs/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_base_finetuned_wikisql_en.md b/docs/_posts/Cabir40/2023-01-30-t5_base_finetuned_wikisql_en.md new file mode 100644 index 00000000000000..e41e50575bfa3b --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_base_finetuned_wikisql_en.md @@ -0,0 +1,93 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from mrm8488) +author: John Snow Labs +name: t5_base_finetuned_wikisql +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-base-finetuned-wikiSQL` is a English model originally trained by `mrm8488`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_base_finetuned_wikisql_en_4.3.0_3.0_1675109286457.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_base_finetuned_wikisql_en_4.3.0_3.0_1675109286457.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_base_finetuned_wikisql","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_base_finetuned_wikisql","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_base_finetuned_wikisql| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|887.6 MB| + +## References + +- https://huggingface.co/mrm8488/t5-base-finetuned-wikiSQL +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://github.com/salesforce/WikiSQL +- https://arxiv.org/pdf/1910.10683.pdf +- https://i.imgur.com/jVFMMWR.png +- https://github.com/patil-suraj/exploring-T5/blob/master/t5_fine_tuning.ipynb +- https://github.com/patil-suraj +- https://pbs.twimg.com/media/Ec5vaG5XsAINty_?format=png&name=900x900 +- https://twitter.com/mrm8488 +- https://www.linkedin.com/in/manuel-romero-cs/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_base_gnad_de.md b/docs/_posts/Cabir40/2023-01-30-t5_base_gnad_de.md new file mode 100644 index 00000000000000..96085660483929 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_base_gnad_de.md @@ -0,0 +1,84 @@ +--- +layout: model +title: German T5ForConditionalGeneration Base Cased model (from Einmalumdiewelt) +author: John Snow Labs +name: t5_base_gnad +date: 2023-01-30 +tags: [de, open_source, t5, tensorflow] +task: Text Generation +language: de +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `T5-Base_GNAD` is a German model originally trained by `Einmalumdiewelt`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_base_gnad_de_4.3.0_3.0_1675099176903.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_base_gnad_de_4.3.0_3.0_1675099176903.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_base_gnad","de") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_base_gnad","de") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_base_gnad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|de| +|Size:|919.7 MB| + +## References + +- https://huggingface.co/Einmalumdiewelt/T5-Base_GNAD \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_base_gnad_maxsamples_de.md b/docs/_posts/Cabir40/2023-01-30-t5_base_gnad_maxsamples_de.md new file mode 100644 index 00000000000000..4b34c3ce80780a --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_base_gnad_maxsamples_de.md @@ -0,0 +1,84 @@ +--- +layout: model +title: German T5ForConditionalGeneration Base Cased model (from Einmalumdiewelt) +author: John Snow Labs +name: t5_base_gnad_maxsamples +date: 2023-01-30 +tags: [de, open_source, t5, tensorflow] +task: Text Generation +language: de +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `T5-Base_GNAD_MaxSamples` is a German model originally trained by `Einmalumdiewelt`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_base_gnad_maxsamples_de_4.3.0_3.0_1675099257674.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_base_gnad_maxsamples_de_4.3.0_3.0_1675099257674.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_base_gnad_maxsamples","de") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_base_gnad_maxsamples","de") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_base_gnad_maxsamples| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|de| +|Size:|922.8 MB| + +## References + +- https://huggingface.co/Einmalumdiewelt/T5-Base_GNAD_MaxSamples \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_base_grammar_correction_en.md b/docs/_posts/Cabir40/2023-01-30-t5_base_grammar_correction_en.md new file mode 100644 index 00000000000000..808afefa952ceb --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_base_grammar_correction_en.md @@ -0,0 +1,87 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from vennify) +author: John Snow Labs +name: t5_base_grammar_correction +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-base-grammar-correction` is a English model originally trained by `vennify`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_base_grammar_correction_en_4.3.0_3.0_1675109588406.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_base_grammar_correction_en_4.3.0_3.0_1675109588406.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_base_grammar_correction","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_base_grammar_correction","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_base_grammar_correction| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|904.6 MB| + +## References + +- https://huggingface.co/vennify/t5-base-grammar-correction +- https://github.com/EricFillion/happy-transformer +- https://arxiv.org/abs/1702.04066 +- https://www.vennify.ai/fine-tune-grammar-correction/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_base_qg_fquad_fr.md b/docs/_posts/Cabir40/2023-01-30-t5_base_qg_fquad_fr.md new file mode 100644 index 00000000000000..9b3085478db948 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_base_qg_fquad_fr.md @@ -0,0 +1,85 @@ +--- +layout: model +title: French T5ForConditionalGeneration Base Cased model (from JDBN) +author: John Snow Labs +name: t5_base_qg_fquad +date: 2023-01-30 +tags: [fr, open_source, t5, tensorflow] +task: Text Generation +language: fr +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-base-fr-qg-fquad` is a French model originally trained by `JDBN`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_base_qg_fquad_fr_4.3.0_3.0_1675109420285.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_base_qg_fquad_fr_4.3.0_3.0_1675109420285.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_base_qg_fquad","fr") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_base_qg_fquad","fr") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_base_qg_fquad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|fr| +|Size:|923.2 MB| + +## References + +- https://huggingface.co/JDBN/t5-base-fr-qg-fquad +- https://github.com/patil-suraj/question_generation \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_base_sum_cnndm_fr.md b/docs/_posts/Cabir40/2023-01-30-t5_base_sum_cnndm_fr.md new file mode 100644 index 00000000000000..e84db87404fad5 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_base_sum_cnndm_fr.md @@ -0,0 +1,84 @@ +--- +layout: model +title: French T5ForConditionalGeneration Base Cased model (from plguillou) +author: John Snow Labs +name: t5_base_sum_cnndm +date: 2023-01-30 +tags: [fr, open_source, t5, tensorflow] +task: Text Generation +language: fr +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-base-fr-sum-cnndm` is a French model originally trained by `plguillou`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_base_sum_cnndm_fr_4.3.0_3.0_1675109501910.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_base_sum_cnndm_fr_4.3.0_3.0_1675109501910.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_base_sum_cnndm","fr") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_base_sum_cnndm","fr") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_base_sum_cnndm| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|fr| +|Size:|923.2 MB| + +## References + +- https://huggingface.co/plguillou/t5-base-fr-sum-cnndm \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_base_v1.1_fine_tuned_for_question_generation_en.md b/docs/_posts/Cabir40/2023-01-30-t5_base_v1.1_fine_tuned_for_question_generation_en.md new file mode 100644 index 00000000000000..d1b8c5cd0b489f --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_base_v1.1_fine_tuned_for_question_generation_en.md @@ -0,0 +1,84 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from ZhangCheng) +author: John Snow Labs +name: t5_base_v1.1_fine_tuned_for_question_generation +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `T5v1.1-Base-Fine-Tuned-for-Question-Generation` is a English model originally trained by `ZhangCheng`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_base_v1.1_fine_tuned_for_question_generation_en_4.3.0_3.0_1675099669115.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_base_v1.1_fine_tuned_for_question_generation_en_4.3.0_3.0_1675099669115.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_base_v1.1_fine_tuned_for_question_generation","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_base_v1.1_fine_tuned_for_question_generation","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_base_v1.1_fine_tuned_for_question_generation| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|1.0 GB| + +## References + +- https://huggingface.co/ZhangCheng/T5v1.1-Base-Fine-Tuned-for-Question-Generation \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_cahya_base_indonesian_summarization_cased_id.md b/docs/_posts/Cabir40/2023-01-30-t5_cahya_base_indonesian_summarization_cased_id.md new file mode 100644 index 00000000000000..d11483eb8045f6 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_cahya_base_indonesian_summarization_cased_id.md @@ -0,0 +1,84 @@ +--- +layout: model +title: Indonesian T5ForConditionalGeneration Base Cased model (from cahya) +author: John Snow Labs +name: t5_cahya_base_indonesian_summarization_cased +date: 2023-01-30 +tags: [id, open_source, t5, tensorflow] +task: Text Generation +language: id +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-base-indonesian-summarization-cased` is a Indonesian model originally trained by `cahya`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_cahya_base_indonesian_summarization_cased_id_4.3.0_3.0_1675109672981.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_cahya_base_indonesian_summarization_cased_id_4.3.0_3.0_1675109672981.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_cahya_base_indonesian_summarization_cased","id") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_cahya_base_indonesian_summarization_cased","id") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_cahya_base_indonesian_summarization_cased| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|id| +|Size:|926.2 MB| + +## References + +- https://huggingface.co/cahya/t5-base-indonesian-summarization-cased \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_caribe_capitalise_en.md b/docs/_posts/Cabir40/2023-01-30-t5_caribe_capitalise_en.md new file mode 100644 index 00000000000000..845f761ee04422 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_caribe_capitalise_en.md @@ -0,0 +1,85 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from KES) +author: John Snow Labs +name: t5_caribe_capitalise +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `caribe-capitalise` is a English model originally trained by `KES`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_caribe_capitalise_en_4.3.0_3.0_1675100357729.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_caribe_capitalise_en_4.3.0_3.0_1675100357729.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_caribe_capitalise","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_caribe_capitalise","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_caribe_capitalise| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|864.8 MB| + +## References + +- https://huggingface.co/KES/caribe-capitalise +- https://pypi.org/project/Caribe/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_chinese_couplet_mengzi_finetune_zh.md b/docs/_posts/Cabir40/2023-01-30-t5_chinese_couplet_mengzi_finetune_zh.md new file mode 100644 index 00000000000000..fe3535ab0ba0e2 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_chinese_couplet_mengzi_finetune_zh.md @@ -0,0 +1,87 @@ +--- +layout: model +title: Chinese T5ForConditionalGeneration Cased model (from hululuzhu) +author: John Snow Labs +name: t5_chinese_couplet_mengzi_finetune +date: 2023-01-30 +tags: [zh, open_source, t5, tensorflow] +task: Text Generation +language: zh +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `chinese-couplet-t5-mengzi-finetune` is a Chinese model originally trained by `hululuzhu`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_chinese_couplet_mengzi_finetune_zh_4.3.0_3.0_1675100437975.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_chinese_couplet_mengzi_finetune_zh_4.3.0_3.0_1675100437975.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_chinese_couplet_mengzi_finetune","zh") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_chinese_couplet_mengzi_finetune","zh") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_chinese_couplet_mengzi_finetune| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|zh| +|Size:|1.0 GB| + +## References + +- https://huggingface.co/hululuzhu/chinese-couplet-t5-mengzi-finetune +- https://github.com/hululuzhu/chinese-ai-writing-share +- https://github.com/hululuzhu/chinese-ai-writing-share/tree/main/slides +- https://github.com/wb14123/couplet-dataset \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_chinese_poem_mengzi_finetune_zh.md b/docs/_posts/Cabir40/2023-01-30-t5_chinese_poem_mengzi_finetune_zh.md new file mode 100644 index 00000000000000..fc764bd5bd3413 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_chinese_poem_mengzi_finetune_zh.md @@ -0,0 +1,87 @@ +--- +layout: model +title: Chinese T5ForConditionalGeneration Cased model (from hululuzhu) +author: John Snow Labs +name: t5_chinese_poem_mengzi_finetune +date: 2023-01-30 +tags: [zh, open_source, t5, tensorflow] +task: Text Generation +language: zh +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `chinese-poem-t5-mengzi-finetune` is a Chinese model originally trained by `hululuzhu`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_chinese_poem_mengzi_finetune_zh_4.3.0_3.0_1675100525010.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_chinese_poem_mengzi_finetune_zh_4.3.0_3.0_1675100525010.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_chinese_poem_mengzi_finetune","zh") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_chinese_poem_mengzi_finetune","zh") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_chinese_poem_mengzi_finetune| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|zh| +|Size:|1.0 GB| + +## References + +- https://huggingface.co/hululuzhu/chinese-poem-t5-mengzi-finetune +- https://github.com/hululuzhu/chinese-ai-writing-share +- https://github.com/hululuzhu/chinese-ai-writing-share/tree/main/slides +- https://github.com/chinese-poetry/chinese-poetry \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_comment_summarization4designtutor_xx.md b/docs/_posts/Cabir40/2023-01-30-t5_comment_summarization4designtutor_xx.md new file mode 100644 index 00000000000000..05c508dbc6972d --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_comment_summarization4designtutor_xx.md @@ -0,0 +1,100 @@ +--- +layout: model +title: Multilingual T5ForConditionalGeneration Cased model (from qiaoyi) +author: John Snow Labs +name: t5_comment_summarization4designtutor +date: 2023-01-30 +tags: [ro, fr, de, en, open_source, t5, xx] +task: Text Generation +language: xx +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `Comment_Summarization4DesignTutor` is a Multilingual model originally trained by `qiaoyi`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_comment_summarization4designtutor_xx_4.3.0_3.0_1675096380129.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_comment_summarization4designtutor_xx_4.3.0_3.0_1675096380129.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols(["text"]) \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_comment_summarization4designtutor","xx") \ + .setInputCols(["document"]) \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_comment_summarization4designtutor","xx") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_comment_summarization4designtutor| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|xx| +|Size:|271.8 MB| + +## References + +- https://huggingface.co/qiaoyi/Comment_Summarization4DesignTutor +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/1805.12471 +- https://nlp.stanford.edu/~socherr/EMNLP2013_RNTN.pdf +- https://aclanthology.org/I05-5002 +- https://arxiv.org/abs/1708.00055 +- https://quoradata.quora.com/First-Quora-Dataset-Release-Question-Pairs +- https://arxiv.org/abs/1704.05426 +- https://arxiv.org/abs/1606.05250 +- https://link.springer.com/chapter/10.1007/11736790_9 +- https://semanticsarchive.net/Archive/Tg3ZGI2M/Marneffe.pdf +- https://www.researchgate.net/publication/221251392_Choice_of_Plausible_Alternatives_An_Evaluation_of_Commonsense_Causal_Reasoning +- https://arxiv.org/abs/1808.09121 +- https://aclanthology.org/N18-1023 +- https://arxiv.org/abs/1810.12885 +- https://arxiv.org/abs/1905.10044 +- https://arxiv.org/pdf/1910.10683.pdf +- https://camo.githubusercontent.com/623b4dea0b653f2ad3f36c71ebfe749a677ac0a1/68747470733a2f2f6d69726f2e6d656469756d2e636f6d2f6d61782f343030362f312a44304a31674e51663876727255704b657944387750412e706e67 diff --git a/docs/_posts/Cabir40/2023-01-30-t5_cover_letter_base_en.md b/docs/_posts/Cabir40/2023-01-30-t5_cover_letter_base_en.md new file mode 100644 index 00000000000000..d6a5e82294a329 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_cover_letter_base_en.md @@ -0,0 +1,84 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from nouamanetazi) +author: John Snow Labs +name: t5_cover_letter_base +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `cover-letter-t5-base` is a English model originally trained by `nouamanetazi`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_cover_letter_base_en_4.3.0_3.0_1675100617268.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_cover_letter_base_en_4.3.0_3.0_1675100617268.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_cover_letter_base","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_cover_letter_base","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_cover_letter_base| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|910.7 MB| + +## References + +- https://huggingface.co/nouamanetazi/cover-letter-t5-base \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_ct5_base_wiki_en.md b/docs/_posts/Cabir40/2023-01-30-t5_ct5_base_wiki_en.md new file mode 100644 index 00000000000000..687d8763e272bb --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_ct5_base_wiki_en.md @@ -0,0 +1,85 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from mtreviso) +author: John Snow Labs +name: t5_ct5_base_wiki +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `ct5-base-en-wiki` is a English model originally trained by `mtreviso`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_ct5_base_wiki_en_4.3.0_3.0_1675100704207.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_ct5_base_wiki_en_4.3.0_3.0_1675100704207.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_ct5_base_wiki","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_ct5_base_wiki","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_ct5_base_wiki| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|928.7 MB| + +## References + +- https://huggingface.co/mtreviso/ct5-base-en-wiki +- https://github.com/mtreviso/chunked-t5 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_ct5_small_wiki_en.md b/docs/_posts/Cabir40/2023-01-30-t5_ct5_small_wiki_en.md new file mode 100644 index 00000000000000..dcd08513c2a6a5 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_ct5_small_wiki_en.md @@ -0,0 +1,85 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from mtreviso) +author: John Snow Labs +name: t5_ct5_small_wiki +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `ct5-small-en-wiki` is a English model originally trained by `mtreviso`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_ct5_small_wiki_en_4.3.0_3.0_1675100765401.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_ct5_small_wiki_en_4.3.0_3.0_1675100765401.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_ct5_small_wiki","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_ct5_small_wiki","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_ct5_small_wiki| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|289.0 MB| + +## References + +- https://huggingface.co/mtreviso/ct5-small-en-wiki +- https://github.com/mtreviso/chunked-t5 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_ct5_small_wiki_l2r_en.md b/docs/_posts/Cabir40/2023-01-30-t5_ct5_small_wiki_l2r_en.md new file mode 100644 index 00000000000000..5014d948e6519e --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_ct5_small_wiki_l2r_en.md @@ -0,0 +1,85 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from mtreviso) +author: John Snow Labs +name: t5_ct5_small_wiki_l2r +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `ct5-small-en-wiki-l2r` is a English model originally trained by `mtreviso`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_ct5_small_wiki_l2r_en_4.3.0_3.0_1675100797165.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_ct5_small_wiki_l2r_en_4.3.0_3.0_1675100797165.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_ct5_small_wiki_l2r","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_ct5_small_wiki_l2r","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_ct5_small_wiki_l2r| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|288.7 MB| + +## References + +- https://huggingface.co/mtreviso/ct5-small-en-wiki-l2r +- https://github.com/mtreviso/chunked-t5 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_diversiformer_de.md b/docs/_posts/Cabir40/2023-01-30-t5_diversiformer_de.md new file mode 100644 index 00000000000000..d319f2e1a2cfa1 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_diversiformer_de.md @@ -0,0 +1,87 @@ +--- +layout: model +title: German T5ForConditionalGeneration Cased model (from diversifix) +author: John Snow Labs +name: t5_diversiformer +date: 2023-01-30 +tags: [de, open_source, t5, tensorflow] +task: Text Generation +language: de +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `diversiformer` is a German model originally trained by `diversifix`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_diversiformer_de_4.3.0_3.0_1675100976411.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_diversiformer_de_4.3.0_3.0_1675100976411.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_diversiformer","de") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_diversiformer","de") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_diversiformer| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|de| +|Size:|1.2 GB| + +## References + +- https://huggingface.co/diversifix/diversiformer +- https://arxiv.org/abs/2010.11934 +- https://github.com/diversifix/diversiformer +- https://www.gnu.org/licenses/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_doc2query_base_msmarco_en.md b/docs/_posts/Cabir40/2023-01-30-t5_doc2query_base_msmarco_en.md new file mode 100644 index 00000000000000..4892041e6c7e1d --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_doc2query_base_msmarco_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from macavaney) +author: John Snow Labs +name: t5_doc2query_base_msmarco +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `doc2query-t5-base-msmarco` is a English model originally trained by `macavaney`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_doc2query_base_msmarco_en_4.3.0_3.0_1675101186297.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_doc2query_base_msmarco_en_4.3.0_3.0_1675101186297.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_doc2query_base_msmarco","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_doc2query_base_msmarco","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_doc2query_base_msmarco| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|474.3 MB| + +## References + +- https://huggingface.co/macavaney/doc2query-t5-base-msmarco +- https://git.uwaterloo.ca/jimmylin/doc2query-data/raw/master/T5-passage/t5-base.zip +- https://github.com/terrierteam/pyterrier_doc2query +- https://cs.uwaterloo.ca/~jimmylin/publications/Nogueira_Lin_2019_docTTTTTquery-v2.pdf +- https://arxiv.org/abs/2007.14271 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_ebanko_base_ru.md b/docs/_posts/Cabir40/2023-01-30-t5_ebanko_base_ru.md new file mode 100644 index 00000000000000..31401835c38239 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_ebanko_base_ru.md @@ -0,0 +1,86 @@ +--- +layout: model +title: Russian T5ForConditionalGeneration Base Cased model (from BlackSamorez) +author: John Snow Labs +name: t5_ebanko_base +date: 2023-01-30 +tags: [ru, open_source, t5, tensorflow] +task: Text Generation +language: ru +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `ebanko-base` is a Russian model originally trained by `BlackSamorez`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_ebanko_base_ru_4.3.0_3.0_1675101255926.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_ebanko_base_ru_4.3.0_3.0_1675101255926.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_ebanko_base","ru") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_ebanko_base","ru") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_ebanko_base| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|ru| +|Size:|927.4 MB| + +## References + +- https://huggingface.co/BlackSamorez/ebanko-base +- https://github.com/BlackSamorez +- https://github.com/skoltech-nlp/russe_detox_2022 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_dl2_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_dl2_en.md new file mode 100644 index 00000000000000..5d8e800eff7e3c --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_dl2_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from google) +author: John Snow Labs +name: t5_efficient_base_dl2 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-base-dl2` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_base_dl2_en_4.3.0_3.0_1675109902657.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_base_dl2_en_4.3.0_3.0_1675109902657.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_base_dl2","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_base_dl2","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_base_dl2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|293.7 MB| + +## References + +- https://huggingface.co/google/t5-efficient-base-dl2 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_dl4_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_dl4_en.md new file mode 100644 index 00000000000000..c3f05073d26098 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_dl4_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from google) +author: John Snow Labs +name: t5_efficient_base_dl4 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-base-dl4` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_base_dl4_en_4.3.0_3.0_1675110033833.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_base_dl4_en_4.3.0_3.0_1675110033833.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_base_dl4","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_base_dl4","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_base_dl4| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|329.9 MB| + +## References + +- https://huggingface.co/google/t5-efficient-base-dl4 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_dl6_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_dl6_en.md new file mode 100644 index 00000000000000..366c972b55491a --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_dl6_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from google) +author: John Snow Labs +name: t5_efficient_base_dl6 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-base-dl6` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_base_dl6_en_4.3.0_3.0_1675110178520.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_base_dl6_en_4.3.0_3.0_1675110178520.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_base_dl6","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_base_dl6","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_base_dl6| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|366.3 MB| + +## References + +- https://huggingface.co/google/t5-efficient-base-dl6 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_dl8_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_dl8_en.md new file mode 100644 index 00000000000000..c068fb603cd371 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_dl8_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from google) +author: John Snow Labs +name: t5_efficient_base_dl8 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-base-dl8` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_base_dl8_en_4.3.0_3.0_1675110336665.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_base_dl8_en_4.3.0_3.0_1675110336665.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_base_dl8","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_base_dl8","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_base_dl8| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|402.3 MB| + +## References + +- https://huggingface.co/google/t5-efficient-base-dl8 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_dm1000_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_dm1000_en.md new file mode 100644 index 00000000000000..b6b0591b0c4412 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_dm1000_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from google) +author: John Snow Labs +name: t5_efficient_base_dm1000 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-base-dm1000` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_base_dm1000_en_4.3.0_3.0_1675110576936.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_base_dm1000_en_4.3.0_3.0_1675110576936.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_base_dm1000","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_base_dm1000","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_base_dm1000| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|633.1 MB| + +## References + +- https://huggingface.co/google/t5-efficient-base-dm1000 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_dm256_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_dm256_en.md new file mode 100644 index 00000000000000..80e491561bc689 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_dm256_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from google) +author: John Snow Labs +name: t5_efficient_base_dm256 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-base-dm256` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_base_dm256_en_4.3.0_3.0_1675110661031.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_base_dm256_en_4.3.0_3.0_1675110661031.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_base_dm256","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_base_dm256","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_base_dm256| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|158.7 MB| + +## References + +- https://huggingface.co/google/t5-efficient-base-dm256 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_dm512_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_dm512_en.md new file mode 100644 index 00000000000000..bad7ad731e1c47 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_dm512_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from google) +author: John Snow Labs +name: t5_efficient_base_dm512 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-base-dm512` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_base_dm512_en_4.3.0_3.0_1675110784924.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_base_dm512_en_4.3.0_3.0_1675110784924.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_base_dm512","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_base_dm512","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_base_dm512| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|316.6 MB| + +## References + +- https://huggingface.co/google/t5-efficient-base-dm512 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_el16_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_el16_en.md new file mode 100644 index 00000000000000..67cbd6f4294198 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_el16_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from google) +author: John Snow Labs +name: t5_efficient_base_el16 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-base-el16` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_base_el16_en_4.3.0_3.0_1675110985082.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_base_el16_en_4.3.0_3.0_1675110985082.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_base_el16","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_base_el16","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_base_el16| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|529.5 MB| + +## References + +- https://huggingface.co/google/t5-efficient-base-el16 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_el2_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_el2_en.md new file mode 100644 index 00000000000000..c81cf3c931a844 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_el2_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from google) +author: John Snow Labs +name: t5_efficient_base_el2 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-base-el2` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_base_el2_en_4.3.0_3.0_1675111126562.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_base_el2_en_4.3.0_3.0_1675111126562.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_base_el2","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_base_el2","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_base_el2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|339.1 MB| + +## References + +- https://huggingface.co/google/t5-efficient-base-el2 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_el4_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_el4_en.md new file mode 100644 index 00000000000000..3b083bca04d7ee --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_el4_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from google) +author: John Snow Labs +name: t5_efficient_base_el4 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-base-el4` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_base_el4_en_4.3.0_3.0_1675111271789.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_base_el4_en_4.3.0_3.0_1675111271789.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_base_el4","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_base_el4","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_base_el4| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|366.4 MB| + +## References + +- https://huggingface.co/google/t5-efficient-base-el4 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_el6_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_el6_en.md new file mode 100644 index 00000000000000..6b5f64e4203edc --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_el6_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from google) +author: John Snow Labs +name: t5_efficient_base_el6 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-base-el6` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_base_el6_en_4.3.0_3.0_1675111427354.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_base_el6_en_4.3.0_3.0_1675111427354.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_base_el6","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_base_el6","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_base_el6| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|393.2 MB| + +## References + +- https://huggingface.co/google/t5-efficient-base-el6 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_el8_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_el8_en.md new file mode 100644 index 00000000000000..4dd3ab75bd6062 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_el8_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from google) +author: John Snow Labs +name: t5_efficient_base_el8 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-base-el8` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_base_el8_en_4.3.0_3.0_1675111594266.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_base_el8_en_4.3.0_3.0_1675111594266.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_base_el8","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_base_el8","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_base_el8| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|420.4 MB| + +## References + +- https://huggingface.co/google/t5-efficient-base-el8 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_ff1000_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_ff1000_en.md new file mode 100644 index 00000000000000..2fe6271e30eeed --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_ff1000_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from google) +author: John Snow Labs +name: t5_efficient_base_ff1000 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-base-ff1000` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_base_ff1000_en_4.3.0_3.0_1675111729938.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_base_ff1000_en_4.3.0_3.0_1675111729938.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_base_ff1000","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_base_ff1000","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_base_ff1000| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|330.1 MB| + +## References + +- https://huggingface.co/google/t5-efficient-base-ff1000 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_ff2000_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_ff2000_en.md new file mode 100644 index 00000000000000..fb0b81572186c7 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_ff2000_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from google) +author: John Snow Labs +name: t5_efficient_base_ff2000 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-base-ff2000` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_base_ff2000_en_4.3.0_3.0_1675111887886.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_base_ff2000_en_4.3.0_3.0_1675111887886.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_base_ff2000","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_base_ff2000","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_base_ff2000| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|402.3 MB| + +## References + +- https://huggingface.co/google/t5-efficient-base-ff2000 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_ff6000_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_ff6000_en.md new file mode 100644 index 00000000000000..6db40b55b7941c --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_ff6000_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from google) +author: John Snow Labs +name: t5_efficient_base_ff6000 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-base-ff6000` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_base_ff6000_en_4.3.0_3.0_1675112144975.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_base_ff6000_en_4.3.0_3.0_1675112144975.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_base_ff6000","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_base_ff6000","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_base_ff6000| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|692.3 MB| + +## References + +- https://huggingface.co/google/t5-efficient-base-ff6000 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_ff9000_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_ff9000_en.md new file mode 100644 index 00000000000000..3d41db9f1877e6 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_ff9000_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from google) +author: John Snow Labs +name: t5_efficient_base_ff9000 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-base-ff9000` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_base_ff9000_en_4.3.0_3.0_1675112485586.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_base_ff9000_en_4.3.0_3.0_1675112485586.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_base_ff9000","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_base_ff9000","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_base_ff9000| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|910.0 MB| + +## References + +- https://huggingface.co/google/t5-efficient-base-ff9000 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_kv128_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_kv128_en.md new file mode 100644 index 00000000000000..73fa68d6ad76bc --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_kv128_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from google) +author: John Snow Labs +name: t5_efficient_base_kv128 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-base-kv128` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_base_kv128_en_4.3.0_3.0_1675112746492.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_base_kv128_en_4.3.0_3.0_1675112746492.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_base_kv128","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_base_kv128","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_base_kv128| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|637.5 MB| + +## References + +- https://huggingface.co/google/t5-efficient-base-kv128 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_nh24_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_nh24_en.md new file mode 100644 index 00000000000000..58bdbd35bb587e --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_nh24_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from google) +author: John Snow Labs +name: t5_efficient_base_nh24 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-base-nh24` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_base_nh24_en_4.3.0_3.0_1675113000788.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_base_nh24_en_4.3.0_3.0_1675113000788.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_base_nh24","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_base_nh24","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_base_nh24| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|636.9 MB| + +## References + +- https://huggingface.co/google/t5-efficient-base-nh24 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_nh32_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_nh32_en.md new file mode 100644 index 00000000000000..27de19fc4ec3bd --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_nh32_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from google) +author: John Snow Labs +name: t5_efficient_base_nh32 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-base-nh32` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_base_nh32_en_4.3.0_3.0_1675113289211.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_base_nh32_en_4.3.0_3.0_1675113289211.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_base_nh32","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_base_nh32","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_base_nh32| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|745.5 MB| + +## References + +- https://huggingface.co/google/t5-efficient-base-nh32 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_nh8_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_nh8_en.md new file mode 100644 index 00000000000000..4b4c4f8435e178 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_nh8_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from google) +author: John Snow Labs +name: t5_efficient_base_nh8 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-base-nh8` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_base_nh8_en_4.3.0_3.0_1675113470480.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_base_nh8_en_4.3.0_3.0_1675113470480.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_base_nh8","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_base_nh8","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_base_nh8| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|420.8 MB| + +## References + +- https://huggingface.co/google/t5-efficient-base-nh8 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_nl16_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_nl16_en.md new file mode 100644 index 00000000000000..907c619da73e99 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_nl16_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from google) +author: John Snow Labs +name: t5_efficient_base_nl16 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-base-nl16` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_base_nl16_en_4.3.0_3.0_1675113699074.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_base_nl16_en_4.3.0_3.0_1675113699074.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_base_nl16","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_base_nl16","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_base_nl16| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|602.0 MB| + +## References + +- https://huggingface.co/google/t5-efficient-base-nl16 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_nl2_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_nl2_en.md new file mode 100644 index 00000000000000..4ce18c1773cb77 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_nl2_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from google) +author: John Snow Labs +name: t5_efficient_base_nl2 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-base-nl2` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_base_nl2_en_4.3.0_3.0_1675113785226.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_base_nl2_en_4.3.0_3.0_1675113785226.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_base_nl2","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_base_nl2","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_base_nl2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|158.2 MB| + +## References + +- https://huggingface.co/google/t5-efficient-base-nl2 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_nl4_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_nl4_en.md new file mode 100644 index 00000000000000..5d853e86c25bd2 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_nl4_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from google) +author: John Snow Labs +name: t5_efficient_base_nl4 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-base-nl4` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_base_nl4_en_4.3.0_3.0_1675113874908.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_base_nl4_en_4.3.0_3.0_1675113874908.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_base_nl4","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_base_nl4","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_base_nl4| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|221.4 MB| + +## References + +- https://huggingface.co/google/t5-efficient-base-nl4 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_nl8_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_nl8_en.md new file mode 100644 index 00000000000000..fe95d165612a4e --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_base_nl8_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from google) +author: John Snow Labs +name: t5_efficient_base_nl8 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-base-nl8` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_base_nl8_en_4.3.0_3.0_1675114011208.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_base_nl8_en_4.3.0_3.0_1675114011208.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_base_nl8","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_base_nl8","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_base_nl8| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|348.0 MB| + +## References + +- https://huggingface.co/google/t5-efficient-base-nl8 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_dl2_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_dl2_en.md new file mode 100644 index 00000000000000..7ec3c304a7bc0a --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_dl2_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Large Cased model (from google) +author: John Snow Labs +name: t5_efficient_large_dl2 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-large-dl2` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_large_dl2_en_4.3.0_3.0_1675114295600.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_large_dl2_en_4.3.0_3.0_1675114295600.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_large_dl2","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_large_dl2","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_large_dl2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|770.5 MB| + +## References + +- https://huggingface.co/google/t5-efficient-large-dl2 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_dl4_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_dl4_en.md new file mode 100644 index 00000000000000..af89b1c9bd4ddb --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_dl4_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Large Cased model (from google) +author: John Snow Labs +name: t5_efficient_large_dl4 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-large-dl4` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_large_dl4_en_4.3.0_3.0_1675114615864.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_large_dl4_en_4.3.0_3.0_1675114615864.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_large_dl4","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_large_dl4","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_large_dl4| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|834.7 MB| + +## References + +- https://huggingface.co/google/t5-efficient-large-dl4 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_dl6_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_dl6_en.md new file mode 100644 index 00000000000000..386b94c84c3945 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_dl6_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Large Cased model (from google) +author: John Snow Labs +name: t5_efficient_large_dl6 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-large-dl6` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_large_dl6_en_4.3.0_3.0_1675114965992.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_large_dl6_en_4.3.0_3.0_1675114965992.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_large_dl6","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_large_dl6","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_large_dl6| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|899.5 MB| + +## References + +- https://huggingface.co/google/t5-efficient-large-dl6 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_dm256_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_dm256_en.md new file mode 100644 index 00000000000000..acf856080c1a4b --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_dm256_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Large Cased model (from google) +author: John Snow Labs +name: t5_efficient_large_dm256 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-large-dm256` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_large_dm256_en_4.3.0_3.0_1675115135612.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_large_dm256_en_4.3.0_3.0_1675115135612.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_large_dm256","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_large_dm256","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_large_dm256| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|370.2 MB| + +## References + +- https://huggingface.co/google/t5-efficient-large-dm256 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_dm512_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_dm512_en.md new file mode 100644 index 00000000000000..a2f4697aca0401 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_dm512_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Large Cased model (from google) +author: John Snow Labs +name: t5_efficient_large_dm512 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-large-dm512` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_large_dm512_en_4.3.0_3.0_1675115410843.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_large_dm512_en_4.3.0_3.0_1675115410843.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_large_dm512","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_large_dm512","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_large_dm512| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|739.8 MB| + +## References + +- https://huggingface.co/google/t5-efficient-large-dm512 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_el2_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_el2_en.md new file mode 100644 index 00000000000000..34271c0902b492 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_el2_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Large Cased model (from google) +author: John Snow Labs +name: t5_efficient_large_el2 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-large-el2` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_large_el2_en_4.3.0_3.0_1675115770479.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_large_el2_en_4.3.0_3.0_1675115770479.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_large_el2","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_large_el2","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_large_el2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|948.2 MB| + +## References + +- https://huggingface.co/google/t5-efficient-large-el2 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_el4_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_el4_en.md new file mode 100644 index 00000000000000..9f1363cc9505b2 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_el4_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Large Cased model (from google) +author: John Snow Labs +name: t5_efficient_large_el4 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-large-el4` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_large_el4_en_4.3.0_3.0_1675116178104.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_large_el4_en_4.3.0_3.0_1675116178104.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_large_el4","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_large_el4","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_large_el4| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|995.5 MB| + +## References + +- https://huggingface.co/google/t5-efficient-large-el4 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_nh2_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_nh2_en.md new file mode 100644 index 00000000000000..4fc0de2163c07e --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_nh2_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Large Cased model (from google) +author: John Snow Labs +name: t5_efficient_large_nh2 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-large-nh2` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_large_nh2_en_4.3.0_3.0_1675116555931.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_large_nh2_en_4.3.0_3.0_1675116555931.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_large_nh2","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_large_nh2","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_large_nh2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|976.3 MB| + +## References + +- https://huggingface.co/google/t5-efficient-large-nh2 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_nl10_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_nl10_en.md new file mode 100644 index 00000000000000..1da0330f24fdb0 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_nl10_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Large Cased model (from google) +author: John Snow Labs +name: t5_efficient_large_nl10 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-large-nl10` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_large_nl10_en_4.3.0_3.0_1675116839759.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_large_nl10_en_4.3.0_3.0_1675116839759.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_large_nl10","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_large_nl10","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_large_nl10| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|689.4 MB| + +## References + +- https://huggingface.co/google/t5-efficient-large-nl10 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_nl12_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_nl12_en.md new file mode 100644 index 00000000000000..4f8c72a3039651 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_nl12_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Large Cased model (from google) +author: John Snow Labs +name: t5_efficient_large_nl12 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-large-nl12` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_large_nl12_en_4.3.0_3.0_1675117150157.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_large_nl12_en_4.3.0_3.0_1675117150157.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_large_nl12","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_large_nl12","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_large_nl12| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|802.0 MB| + +## References + +- https://huggingface.co/google/t5-efficient-large-nl12 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_nl16_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_nl16_en.md new file mode 100644 index 00000000000000..db79fd5589b1e4 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_nl16_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Large Cased model (from google) +author: John Snow Labs +name: t5_efficient_large_nl16 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-large-nl16` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_large_nl16_en_4.3.0_3.0_1675117543768.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_large_nl16_en_4.3.0_3.0_1675117543768.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_large_nl16","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_large_nl16","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_large_nl16| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|1.0 GB| + +## References + +- https://huggingface.co/google/t5-efficient-large-nl16 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_nl2_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_nl2_en.md new file mode 100644 index 00000000000000..7130d99cf72fa5 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_nl2_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Large Cased model (from google) +author: John Snow Labs +name: t5_efficient_large_nl2 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-large-nl2` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_large_nl2_en_4.3.0_3.0_1675117675037.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_large_nl2_en_4.3.0_3.0_1675117675037.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_large_nl2","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_large_nl2","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_large_nl2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|238.9 MB| + +## References + +- https://huggingface.co/google/t5-efficient-large-nl2 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_nl4_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_nl4_en.md new file mode 100644 index 00000000000000..f651591ddf769a --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_nl4_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Large Cased model (from google) +author: John Snow Labs +name: t5_efficient_large_nl4 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-large-nl4` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_large_nl4_en_4.3.0_3.0_1675117814381.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_large_nl4_en_4.3.0_3.0_1675117814381.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_large_nl4","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_large_nl4","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_large_nl4| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|351.1 MB| + +## References + +- https://huggingface.co/google/t5-efficient-large-nl4 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_nl8_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_nl8_en.md new file mode 100644 index 00000000000000..5277827c7c71a2 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_large_nl8_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Large Cased model (from google) +author: John Snow Labs +name: t5_efficient_large_nl8 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-large-nl8` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_large_nl8_en_4.3.0_3.0_1675118058956.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_large_nl8_en_4.3.0_3.0_1675118058956.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_large_nl8","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_large_nl8","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_large_nl8| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|576.4 MB| + +## References + +- https://huggingface.co/google/t5-efficient-large-nl8 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_mini_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_mini_en.md new file mode 100644 index 00000000000000..23e3e0130f30fa --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_mini_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Mini Cased model (from google) +author: John Snow Labs +name: t5_efficient_mini +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-mini` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_mini_en_4.3.0_3.0_1675118119036.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_mini_en_4.3.0_3.0_1675118119036.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_mini","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_mini","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_mini| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|83.9 MB| + +## References + +- https://huggingface.co/google/t5-efficient-mini +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_mini_nl12_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_mini_nl12_en.md new file mode 100644 index 00000000000000..0350ef59497f39 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_mini_nl12_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Mini Cased model (from google) +author: John Snow Labs +name: t5_efficient_mini_nl12 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-mini-nl12` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_mini_nl12_en_4.3.0_3.0_1675118188776.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_mini_nl12_en_4.3.0_3.0_1675118188776.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_mini_nl12","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_mini_nl12","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_mini_nl12| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|156.4 MB| + +## References + +- https://huggingface.co/google/t5-efficient-mini-nl12 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_mini_nl24_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_mini_nl24_en.md new file mode 100644 index 00000000000000..022b84ff368b7f --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_mini_nl24_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Mini Cased model (from google) +author: John Snow Labs +name: t5_efficient_mini_nl24 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-mini-nl24` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_mini_nl24_en_4.3.0_3.0_1675118296258.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_mini_nl24_en_4.3.0_3.0_1675118296258.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_mini_nl24","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_mini_nl24","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_mini_nl24| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|265.3 MB| + +## References + +- https://huggingface.co/google/t5-efficient-mini-nl24 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_mini_nl6_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_mini_nl6_en.md new file mode 100644 index 00000000000000..82d3211306ed08 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_mini_nl6_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Mini Cased model (from google) +author: John Snow Labs +name: t5_efficient_mini_nl6 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-mini-nl6` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_mini_nl6_en_4.3.0_3.0_1675118348456.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_mini_nl6_en_4.3.0_3.0_1675118348456.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_mini_nl6","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_mini_nl6","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_mini_nl6| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|102.0 MB| + +## References + +- https://huggingface.co/google/t5-efficient-mini-nl6 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_mini_nl8_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_mini_nl8_en.md new file mode 100644 index 00000000000000..465429a88b544e --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_mini_nl8_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Mini Cased model (from google) +author: John Snow Labs +name: t5_efficient_mini_nl8 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-mini-nl8` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_mini_nl8_en_4.3.0_3.0_1675118402070.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_mini_nl8_en_4.3.0_3.0_1675118402070.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_mini_nl8","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_mini_nl8","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_mini_nl8| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|120.1 MB| + +## References + +- https://huggingface.co/google/t5-efficient-mini-nl8 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_dl12_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_dl12_en.md new file mode 100644 index 00000000000000..9511cdcb6b2d05 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_dl12_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_dl12 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-dl12` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_dl12_en_4.3.0_3.0_1675118545379.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_dl12_en_4.3.0_3.0_1675118545379.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_dl12","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_dl12","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_dl12| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|196.2 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-dl12 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_dl16_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_dl16_en.md new file mode 100644 index 00000000000000..e08866ac65b0b9 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_dl16_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_dl16 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-dl16` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_dl16_en_4.3.0_3.0_1675118638338.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_dl16_en_4.3.0_3.0_1675118638338.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_dl16","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_dl16","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_dl16| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|228.3 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-dl16 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_dl2_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_dl2_en.md new file mode 100644 index 00000000000000..0b477d09f236c8 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_dl2_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_dl2 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-dl2` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_dl2_en_4.3.0_3.0_1675118693652.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_dl2_en_4.3.0_3.0_1675118693652.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_dl2","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_dl2","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_dl2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|115.7 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-dl2 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_dl4_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_dl4_en.md new file mode 100644 index 00000000000000..bfa8ba4ed56adb --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_dl4_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_dl4 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-dl4` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_dl4_en_4.3.0_3.0_1675118749298.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_dl4_en_4.3.0_3.0_1675118749298.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_dl4","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_dl4","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_dl4| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|131.9 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-dl4 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_dl8_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_dl8_en.md new file mode 100644 index 00000000000000..ac64502024c1ed --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_dl8_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_dl8 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-dl8` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_dl8_en_4.3.0_3.0_1675118817239.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_dl8_en_4.3.0_3.0_1675118817239.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_dl8","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_dl8","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_dl8| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|164.0 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-dl8 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_dm1000_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_dm1000_en.md new file mode 100644 index 00000000000000..ba7af47b759c81 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_dm1000_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_dm1000 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-dm1000` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_dm1000_en_4.3.0_3.0_1675118935697.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_dm1000_en_4.3.0_3.0_1675118935697.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_dm1000","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_dm1000","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_dm1000| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|295.2 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-dm1000 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_dm128_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_dm128_en.md new file mode 100644 index 00000000000000..1b05b64787c303 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_dm128_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_dm128 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-dm128` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_dm128_en_4.3.0_3.0_1675118965696.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_dm128_en_4.3.0_3.0_1675118965696.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_dm128","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_dm128","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_dm128| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|37.4 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-dm128 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_dm2000_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_dm2000_en.md new file mode 100644 index 00000000000000..aa4d1d16fc3dba --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_dm2000_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_dm2000 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-dm2000` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_dm2000_en_4.3.0_3.0_1675119184723.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_dm2000_en_4.3.0_3.0_1675119184723.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_dm2000","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_dm2000","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_dm2000| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|590.0 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-dm2000 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_dm256_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_dm256_en.md new file mode 100644 index 00000000000000..2943cf61547329 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_dm256_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_dm256 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-dm256` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_dm256_en_4.3.0_3.0_1675119242272.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_dm256_en_4.3.0_3.0_1675119242272.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_dm256","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_dm256","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_dm256| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|74.2 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-dm256 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_dm768_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_dm768_en.md new file mode 100644 index 00000000000000..2727bce4667f3a --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_dm768_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_dm768 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-dm768` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_dm768_en_4.3.0_3.0_1675119328836.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_dm768_en_4.3.0_3.0_1675119328836.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_dm768","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_dm768","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_dm768| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|221.5 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-dm768 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el12_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el12_en.md new file mode 100644 index 00000000000000..346f6e48f746cc --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el12_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_el12 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-el12` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_el12_en_4.3.0_3.0_1675119407178.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_el12_en_4.3.0_3.0_1675119407178.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_el12","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_el12","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_el12| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|184.0 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-el12 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el16_dl1_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el16_dl1_en.md new file mode 100644 index 00000000000000..16ec63ee062d7b --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el16_dl1_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_el16_dl1 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-el16-dl1` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_el16_dl1_en_4.3.0_3.0_1675119566747.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_el16_dl1_en_4.3.0_3.0_1675119566747.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_el16_dl1","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_el16_dl1","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_el16_dl1| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|168.0 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-el16-dl1 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el16_dl2_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el16_dl2_en.md new file mode 100644 index 00000000000000..87d51026a45d2c --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el16_dl2_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_el16_dl2 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-el16-dl2` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_el16_dl2_en_4.3.0_3.0_1675119642706.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_el16_dl2_en_4.3.0_3.0_1675119642706.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_el16_dl2","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_el16_dl2","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_el16_dl2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|175.9 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-el16-dl2 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el16_dl4_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el16_dl4_en.md new file mode 100644 index 00000000000000..32b8d4491a2dba --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el16_dl4_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_el16_dl4 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-el16-dl4` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_el16_dl4_en_4.3.0_3.0_1675119722159.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_el16_dl4_en_4.3.0_3.0_1675119722159.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_el16_dl4","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_el16_dl4","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_el16_dl4| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|192.1 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-el16-dl4 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el16_dl8_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el16_dl8_en.md new file mode 100644 index 00000000000000..ba3e92f1909ffe --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el16_dl8_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_el16_dl8 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-el16-dl8` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_el16_dl8_en_4.3.0_3.0_1675119813436.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_el16_dl8_en_4.3.0_3.0_1675119813436.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_el16_dl8","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_el16_dl8","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_el16_dl8| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|224.3 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-el16-dl8 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el16_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el16_en.md new file mode 100644 index 00000000000000..ad85ba29efff79 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el16_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_el16 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-el16` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_el16_en_4.3.0_3.0_1675119492656.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_el16_en_4.3.0_3.0_1675119492656.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_el16","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_el16","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_el16| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|208.2 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-el16 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el2_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el2_en.md new file mode 100644 index 00000000000000..771677d2e4d502 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el2_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_el2 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-el2` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_el2_en_4.3.0_3.0_1675119873194.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_el2_en_4.3.0_3.0_1675119873194.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_el2","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_el2","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_el2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|123.8 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-el2 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el32_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el32_en.md new file mode 100644 index 00000000000000..e9b2fabd31bf6d --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el32_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_el32 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-el32` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_el32_en_4.3.0_3.0_1675120001777.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_el32_en_4.3.0_3.0_1675120001777.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_el32","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_el32","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_el32| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|305.0 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-el32 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el48_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el48_en.md new file mode 100644 index 00000000000000..fa55ea8cd47c42 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el48_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_el48 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-el48` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_el48_en_4.3.0_3.0_1675120232514.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_el48_en_4.3.0_3.0_1675120232514.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_el48","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_el48","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_el48| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|401.7 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-el48 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el4_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el4_en.md new file mode 100644 index 00000000000000..b97b7e2e91d075 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el4_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_el4 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-el4` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_el4_en_4.3.0_3.0_1675120069644.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_el4_en_4.3.0_3.0_1675120069644.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_el4","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_el4","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_el4| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|135.8 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-el4 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el64_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el64_en.md new file mode 100644 index 00000000000000..4cc91498b9ae40 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el64_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_el64 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-el64` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_el64_en_4.3.0_3.0_1675120423638.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_el64_en_4.3.0_3.0_1675120423638.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_el64","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_el64","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_el64| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|499.3 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-el64 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el8_dl1_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el8_dl1_en.md new file mode 100644 index 00000000000000..7a1a17c6871a8d --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el8_dl1_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_el8_dl1 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-el8-dl1` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_el8_dl1_en_4.3.0_3.0_1675120557059.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_el8_dl1_en_4.3.0_3.0_1675120557059.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_el8_dl1","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_el8_dl1","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_el8_dl1| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|119.7 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-el8-dl1 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el8_dl2_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el8_dl2_en.md new file mode 100644 index 00000000000000..c31ff863ab035d --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el8_dl2_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_el8_dl2 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-el8-dl2` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_el8_dl2_en_4.3.0_3.0_1675120612912.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_el8_dl2_en_4.3.0_3.0_1675120612912.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_el8_dl2","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_el8_dl2","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_el8_dl2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|127.7 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-el8-dl2 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el8_dl4_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el8_dl4_en.md new file mode 100644 index 00000000000000..cc2c1333d06463 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el8_dl4_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_el8_dl4 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-el8-dl4` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_el8_dl4_en_4.3.0_3.0_1675120674389.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_el8_dl4_en_4.3.0_3.0_1675120674389.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_el8_dl4","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_el8_dl4","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_el8_dl4| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|143.8 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-el8-dl4 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el8_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el8_en.md new file mode 100644 index 00000000000000..bad69c30f5060d --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_el8_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_el8 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-el8` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_el8_en_4.3.0_3.0_1675120502722.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_el8_en_4.3.0_3.0_1675120502722.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_el8","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_el8","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_el8| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|159.9 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-el8 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_en.md new file mode 100644 index 00000000000000..4bdea8c0d907f5 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_en_4.3.0_3.0_1675118464936.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_en_4.3.0_3.0_1675118464936.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|147.8 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_ff1000_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_ff1000_en.md new file mode 100644 index 00000000000000..8bc366a4f9cf62 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_ff1000_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_ff1000 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-ff1000` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_ff1000_en_4.3.0_3.0_1675120730150.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_ff1000_en_4.3.0_3.0_1675120730150.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_ff1000","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_ff1000","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_ff1000| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|123.8 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-ff1000 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_ff12000_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_ff12000_en.md new file mode 100644 index 00000000000000..e6c4f79af31e13 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_ff12000_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_ff12000 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-ff12000` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_ff12000_en_4.3.0_3.0_1675120874161.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_ff12000_en_4.3.0_3.0_1675120874161.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_ff12000","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_ff12000","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_ff12000| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|388.5 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-ff12000 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_ff3000_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_ff3000_en.md new file mode 100644 index 00000000000000..806d77c92fd756 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_ff3000_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_ff3000 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-ff3000` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_ff3000_en_4.3.0_3.0_1675120955500.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_ff3000_en_4.3.0_3.0_1675120955500.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_ff3000","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_ff3000","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_ff3000| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|171.8 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-ff3000 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_ff6000_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_ff6000_en.md new file mode 100644 index 00000000000000..6ab6bc93ec4a99 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_ff6000_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_ff6000 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-ff6000` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_ff6000_en_4.3.0_3.0_1675121052795.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_ff6000_en_4.3.0_3.0_1675121052795.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_ff6000","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_ff6000","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_ff6000| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|244.0 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-ff6000 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_ff9000_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_ff9000_en.md new file mode 100644 index 00000000000000..53ee4b3273abd4 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_ff9000_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_ff9000 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-ff9000` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_ff9000_en_4.3.0_3.0_1675121175904.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_ff9000_en_4.3.0_3.0_1675121175904.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_ff9000","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_ff9000","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_ff9000| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|316.3 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-ff9000 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_kv128_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_kv128_en.md new file mode 100644 index 00000000000000..cb39bbaf604b28 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_kv128_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_kv128 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-kv128` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_kv128_en_4.3.0_3.0_1675121257275.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_kv128_en_4.3.0_3.0_1675121257275.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_kv128","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_kv128","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_kv128| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|183.9 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-kv128 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_kv16_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_kv16_en.md new file mode 100644 index 00000000000000..787bdde3a06216 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_kv16_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_kv16 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-kv16` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_kv16_en_4.3.0_3.0_1675121314422.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_kv16_en_4.3.0_3.0_1675121314422.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_kv16","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_kv16","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_kv16| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|120.7 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-kv16 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_kv256_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_kv256_en.md new file mode 100644 index 00000000000000..b74133a0719aa2 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_kv256_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_kv256 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-kv256` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_kv256_en_4.3.0_3.0_1675121414682.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_kv256_en_4.3.0_3.0_1675121414682.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_kv256","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_kv256","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_kv256| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|256.0 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-kv256 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_kv32_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_kv32_en.md new file mode 100644 index 00000000000000..92ac52ebf0a60e --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_kv32_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_kv32 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-kv32` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_kv32_en_4.3.0_3.0_1675121475776.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_kv32_en_4.3.0_3.0_1675121475776.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_kv32","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_kv32","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_kv32| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|129.8 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-kv32 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_nl16_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_nl16_en.md new file mode 100644 index 00000000000000..b19a633e7789d4 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_nl16_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_nl16 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-nl16` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_nl16_en_4.3.0_3.0_1675121587021.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_nl16_en_4.3.0_3.0_1675121587021.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_nl16","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_nl16","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_nl16| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|288.9 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-nl16 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_nl20_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_nl20_en.md new file mode 100644 index 00000000000000..069c6f7e6787d4 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_nl20_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_nl20 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-nl20` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_nl20_en_4.3.0_3.0_1675121765443.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_nl20_en_4.3.0_3.0_1675121765443.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_nl20","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_nl20","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_nl20| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|345.3 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-nl20 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_nl22_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_nl22_en.md new file mode 100644 index 00000000000000..30c182ba5f0c89 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_nl22_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_nl22 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-nl22` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_nl22_en_4.3.0_3.0_1675121912411.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_nl22_en_4.3.0_3.0_1675121912411.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_nl22","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_nl22","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_nl22| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|373.8 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-nl22 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_nl24_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_nl24_en.md new file mode 100644 index 00000000000000..dbb0afadaefbc3 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_nl24_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_nl24 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-nl24` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_nl24_en_4.3.0_3.0_1675122070674.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_nl24_en_4.3.0_3.0_1675122070674.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_nl24","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_nl24","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_nl24| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|402.2 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-nl24 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_nl2_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_nl2_en.md new file mode 100644 index 00000000000000..467a63edde4ec9 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_nl2_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_nl2 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-nl2` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_nl2_en_4.3.0_3.0_1675121636200.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_nl2_en_4.3.0_3.0_1675121636200.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_nl2","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_nl2","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_nl2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|91.6 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-nl2 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_nl32_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_nl32_en.md new file mode 100644 index 00000000000000..548e678455c16d --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_nl32_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_nl32 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-nl32` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_nl32_en_4.3.0_3.0_1675122269324.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_nl32_en_4.3.0_3.0_1675122269324.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_nl32","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_nl32","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_nl32| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|514.9 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-nl32 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_nl36_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_nl36_en.md new file mode 100644 index 00000000000000..cd688c49f66ea2 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_nl36_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_nl36 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-nl36` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_nl36_en_4.3.0_3.0_1675122498742.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_nl36_en_4.3.0_3.0_1675122498742.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_nl36","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_nl36","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_nl36| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|572.2 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-nl36 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_nl40_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_nl40_en.md new file mode 100644 index 00000000000000..c5eb779bf29da0 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_nl40_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_nl40 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-nl40` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_nl40_en_4.3.0_3.0_1675122791666.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_nl40_en_4.3.0_3.0_1675122791666.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_nl40","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_nl40","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_nl40| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|628.5 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-nl40 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_nl48_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_nl48_en.md new file mode 100644 index 00000000000000..018212450689a0 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_nl48_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_nl48 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-nl48` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_nl48_en_4.3.0_3.0_1675123074393.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_nl48_en_4.3.0_3.0_1675123074393.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_nl48","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_nl48","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_nl48| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|742.0 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-nl48 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_nl4_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_nl4_en.md new file mode 100644 index 00000000000000..054e5ae2a0fa75 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_nl4_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_nl4 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-nl4` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_nl4_en_4.3.0_3.0_1675122566099.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_nl4_en_4.3.0_3.0_1675122566099.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_nl4","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_nl4","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_nl4| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|119.7 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-nl4 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_nl8_en.md b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_nl8_en.md new file mode 100644 index 00000000000000..86c33d5c3ffcca --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_efficient_small_nl8_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_efficient_small_nl8 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-small-nl8` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_small_nl8_en_4.3.0_3.0_1675123172323.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_small_nl8_en_4.3.0_3.0_1675123172323.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_small_nl8","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_small_nl8","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_small_nl8| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|176.0 MB| + +## References + +- https://huggingface.co/google/t5-efficient-small-nl8 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_envit5_translation_xx.md b/docs/_posts/Cabir40/2023-01-30-t5_envit5_translation_xx.md new file mode 100644 index 00000000000000..2a655be0dbe740 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_envit5_translation_xx.md @@ -0,0 +1,89 @@ +--- +layout: model +title: Multilingual T5ForConditionalGeneration Cased model (from VietAI) +author: John Snow Labs +name: t5_envit5_translation +date: 2023-01-30 +tags: [vi, en, open_source, t5, xx, tensorflow] +task: Text Generation +language: xx +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `envit5-translation` is a Multilingual model originally trained by `VietAI`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_envit5_translation_xx_4.3.0_3.0_1675101501844.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_envit5_translation_xx_4.3.0_3.0_1675101501844.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_envit5_translation","xx") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_envit5_translation","xx") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_envit5_translation| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|xx| +|Size:|599.3 MB| + +## References + +- https://huggingface.co/VietAI/envit5-translation +- https://paperswithcode.com/sota/machine-translation-on-iwslt2015-english-1?p=mtet-multi-domain-translation-for-english +- https://paperswithcode.com/sota/on-phomt?p=mtet-multi-domain-translation-for-english-and +- https://research.vietai.org/mtet/ +- https://github.com/VinAIResearch/PhoMT +- https://user-images.githubusercontent.com/44376091/195998681-5860e443-2071-4048-8a2b-873dcee14a72.png \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_est5base_es.md b/docs/_posts/Cabir40/2023-01-30-t5_est5base_es.md new file mode 100644 index 00000000000000..43890f2dfe9347 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_est5base_es.md @@ -0,0 +1,85 @@ +--- +layout: model +title: Spanish T5ForConditionalGeneration Cased model (from JorgeSarry) +author: John Snow Labs +name: t5_est5base +date: 2023-01-30 +tags: [es, open_source, t5, tensorflow] +task: Text Generation +language: es +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `est5base` is a Spanish model originally trained by `JorgeSarry`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_est5base_es_4.3.0_3.0_1675101719578.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_est5base_es_4.3.0_3.0_1675101719578.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_est5base","es") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_est5base","es") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_est5base| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|es| +|Size:|511.9 MB| + +## References + +- https://huggingface.co/JorgeSarry/est5base +- https://towardsdatascience.com/how-to-adapt-a-multilingual-t5-model-for-a-single-language-b9f94f3d9c90 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_eva_forum_headlines_ru.md b/docs/_posts/Cabir40/2023-01-30-t5_eva_forum_headlines_ru.md new file mode 100644 index 00000000000000..516e11f15605ee --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_eva_forum_headlines_ru.md @@ -0,0 +1,85 @@ +--- +layout: model +title: Russian T5ForConditionalGeneration Cased model (from Kateryna) +author: John Snow Labs +name: t5_eva_forum_headlines +date: 2023-01-30 +tags: [ru, open_source, t5, tensorflow] +task: Text Generation +language: ru +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `eva_ru_forum_headlines` is a Russian model originally trained by `Kateryna`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_eva_forum_headlines_ru_4.3.0_3.0_1675101795155.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_eva_forum_headlines_ru_4.3.0_3.0_1675101795155.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_eva_forum_headlines","ru") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_eva_forum_headlines","ru") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_eva_forum_headlines| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|ru| +|Size:|981.6 MB| + +## References + +- https://huggingface.co/Kateryna/eva_ru_forum_headlines +- https://github.com/KaterynaD/eva.ru/tree/main/Code/Notebooks/9.%20Headlines \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_fake_news_detector_en.md b/docs/_posts/Cabir40/2023-01-30-t5_fake_news_detector_en.md new file mode 100644 index 00000000000000..57fab5f7c3f3da --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_fake_news_detector_en.md @@ -0,0 +1,85 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from cometrain) +author: John Snow Labs +name: t5_fake_news_detector +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `fake-news-detector-t5` is a English model originally trained by `cometrain`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_fake_news_detector_en_4.3.0_3.0_1675101857981.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_fake_news_detector_en_4.3.0_3.0_1675101857981.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_fake_news_detector","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_fake_news_detector","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_fake_news_detector| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|277.3 MB| + +## References + +- https://huggingface.co/cometrain/fake-news-detector-t5 +- https://www.kaggle.com/datasets/clmentbisaillon/fake-and-real-news-dataset \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_fine_tuned_model_en.md b/docs/_posts/Cabir40/2023-01-30-t5_fine_tuned_model_en.md new file mode 100644 index 00000000000000..6dc0912f152ad5 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_fine_tuned_model_en.md @@ -0,0 +1,85 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from marcus2000) +author: John Snow Labs +name: t5_fine_tuned_model +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `fine_tuned_t5_model` is a English model originally trained by `marcus2000`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_fine_tuned_model_en_4.3.0_3.0_1675101922079.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_fine_tuned_model_en_4.3.0_3.0_1675101922079.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_fine_tuned_model","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_fine_tuned_model","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_fine_tuned_model| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|918.7 MB| + +## References + +- https://huggingface.co/marcus2000/fine_tuned_t5_model +- https://paperswithcode.com/sota?task=automatic-speech-recognition&dataset=Librispeech+%28clean%29 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_finetune_paraphrase_base_standard_bahasa_cased_ms.md b/docs/_posts/Cabir40/2023-01-30-t5_finetune_paraphrase_base_standard_bahasa_cased_ms.md new file mode 100644 index 00000000000000..03493be07a301e --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_finetune_paraphrase_base_standard_bahasa_cased_ms.md @@ -0,0 +1,85 @@ +--- +layout: model +title: Malay T5ForConditionalGeneration Base Cased model (from mesolitica) +author: John Snow Labs +name: t5_finetune_paraphrase_base_standard_bahasa_cased +date: 2023-01-30 +tags: [ms, open_source, t5, tensorflow] +task: Text Generation +language: ms +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `finetune-paraphrase-t5-base-standard-bahasa-cased` is a Malay model originally trained by `mesolitica`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_finetune_paraphrase_base_standard_bahasa_cased_ms_4.3.0_3.0_1675102005289.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_finetune_paraphrase_base_standard_bahasa_cased_ms_4.3.0_3.0_1675102005289.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_finetune_paraphrase_base_standard_bahasa_cased","ms") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_finetune_paraphrase_base_standard_bahasa_cased","ms") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_finetune_paraphrase_base_standard_bahasa_cased| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|ms| +|Size:|926.8 MB| + +## References + +- https://huggingface.co/mesolitica/finetune-paraphrase-t5-base-standard-bahasa-cased +- https://github.com/huseinzol05/malaya/tree/master/session/paraphrase/hf-t5 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_finetune_paraphrase_small_standard_bahasa_cased_ms.md b/docs/_posts/Cabir40/2023-01-30-t5_finetune_paraphrase_small_standard_bahasa_cased_ms.md new file mode 100644 index 00000000000000..540d70a7f51f0c --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_finetune_paraphrase_small_standard_bahasa_cased_ms.md @@ -0,0 +1,85 @@ +--- +layout: model +title: Malay T5ForConditionalGeneration Small Cased model (from mesolitica) +author: John Snow Labs +name: t5_finetune_paraphrase_small_standard_bahasa_cased +date: 2023-01-30 +tags: [ms, open_source, t5, tensorflow] +task: Text Generation +language: ms +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `finetune-paraphrase-t5-small-standard-bahasa-cased` is a Malay model originally trained by `mesolitica`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_finetune_paraphrase_small_standard_bahasa_cased_ms_4.3.0_3.0_1675102064396.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_finetune_paraphrase_small_standard_bahasa_cased_ms_4.3.0_3.0_1675102064396.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_finetune_paraphrase_small_standard_bahasa_cased","ms") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_finetune_paraphrase_small_standard_bahasa_cased","ms") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_finetune_paraphrase_small_standard_bahasa_cased| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|ms| +|Size:|288.6 MB| + +## References + +- https://huggingface.co/mesolitica/finetune-paraphrase-t5-small-standard-bahasa-cased +- https://github.com/huseinzol05/malaya/tree/master/session/paraphrase/hf-t5 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_finetune_paraphrase_tiny_standard_bahasa_cased_ms.md b/docs/_posts/Cabir40/2023-01-30-t5_finetune_paraphrase_tiny_standard_bahasa_cased_ms.md new file mode 100644 index 00000000000000..8186b5d2a780f9 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_finetune_paraphrase_tiny_standard_bahasa_cased_ms.md @@ -0,0 +1,85 @@ +--- +layout: model +title: Malay T5ForConditionalGeneration Tiny Cased model (from mesolitica) +author: John Snow Labs +name: t5_finetune_paraphrase_tiny_standard_bahasa_cased +date: 2023-01-30 +tags: [ms, open_source, t5, tensorflow] +task: Text Generation +language: ms +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `finetune-paraphrase-t5-tiny-standard-bahasa-cased` is a Malay model originally trained by `mesolitica`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_finetune_paraphrase_tiny_standard_bahasa_cased_ms_4.3.0_3.0_1675102090742.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_finetune_paraphrase_tiny_standard_bahasa_cased_ms_4.3.0_3.0_1675102090742.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_finetune_paraphrase_tiny_standard_bahasa_cased","ms") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_finetune_paraphrase_tiny_standard_bahasa_cased","ms") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_finetune_paraphrase_tiny_standard_bahasa_cased| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|ms| +|Size:|176.9 MB| + +## References + +- https://huggingface.co/mesolitica/finetune-paraphrase-t5-tiny-standard-bahasa-cased +- https://github.com/huseinzol05/malaya/tree/master/session/paraphrase/hf-t5 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_finetune_translation_base_standard_bahasa_cased_ms.md b/docs/_posts/Cabir40/2023-01-30-t5_finetune_translation_base_standard_bahasa_cased_ms.md new file mode 100644 index 00000000000000..7cb57774999106 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_finetune_translation_base_standard_bahasa_cased_ms.md @@ -0,0 +1,86 @@ +--- +layout: model +title: Malay T5ForConditionalGeneration Base Cased model (from mesolitica) +author: John Snow Labs +name: t5_finetune_translation_base_standard_bahasa_cased +date: 2023-01-30 +tags: [ms, open_source, t5, tensorflow] +task: Text Generation +language: ms +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `finetune-translation-t5-base-standard-bahasa-cased` is a Malay model originally trained by `mesolitica`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_finetune_translation_base_standard_bahasa_cased_ms_4.3.0_3.0_1675102146994.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_finetune_translation_base_standard_bahasa_cased_ms_4.3.0_3.0_1675102146994.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_finetune_translation_base_standard_bahasa_cased","ms") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_finetune_translation_base_standard_bahasa_cased","ms") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_finetune_translation_base_standard_bahasa_cased| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|ms| +|Size:|928.4 MB| + +## References + +- https://huggingface.co/mesolitica/finetune-translation-t5-base-standard-bahasa-cased +- https://github.com/huseinzol05/malay-dataset/tree/master/translation/laser +- https://github.com/huseinzol05/malaya/tree/master/session/translation/hf-t5 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_finetune_translation_small_standard_bahasa_cased_ms.md b/docs/_posts/Cabir40/2023-01-30-t5_finetune_translation_small_standard_bahasa_cased_ms.md new file mode 100644 index 00000000000000..bc00f3006d9b03 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_finetune_translation_small_standard_bahasa_cased_ms.md @@ -0,0 +1,86 @@ +--- +layout: model +title: Malay T5ForConditionalGeneration Small Cased model (from mesolitica) +author: John Snow Labs +name: t5_finetune_translation_small_standard_bahasa_cased +date: 2023-01-30 +tags: [ms, open_source, t5, tensorflow] +task: Text Generation +language: ms +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `finetune-translation-t5-small-standard-bahasa-cased` is a Malay model originally trained by `mesolitica`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_finetune_translation_small_standard_bahasa_cased_ms_4.3.0_3.0_1675102196206.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_finetune_translation_small_standard_bahasa_cased_ms_4.3.0_3.0_1675102196206.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_finetune_translation_small_standard_bahasa_cased","ms") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_finetune_translation_small_standard_bahasa_cased","ms") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_finetune_translation_small_standard_bahasa_cased| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|ms| +|Size:|288.7 MB| + +## References + +- https://huggingface.co/mesolitica/finetune-translation-t5-small-standard-bahasa-cased +- https://github.com/huseinzol05/malay-dataset/tree/master/translation/laser +- https://github.com/huseinzol05/malaya/tree/master/session/translation/hf-t5 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_finetune_translation_super_tiny_standard_bahasa_cased_ms.md b/docs/_posts/Cabir40/2023-01-30-t5_finetune_translation_super_tiny_standard_bahasa_cased_ms.md new file mode 100644 index 00000000000000..c630802d3ffa6b --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_finetune_translation_super_tiny_standard_bahasa_cased_ms.md @@ -0,0 +1,86 @@ +--- +layout: model +title: Malay T5ForConditionalGeneration Tiny Cased model (from mesolitica) +author: John Snow Labs +name: t5_finetune_translation_super_tiny_standard_bahasa_cased +date: 2023-01-30 +tags: [ms, open_source, t5, tensorflow] +task: Text Generation +language: ms +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `finetune-translation-t5-super-super-tiny-standard-bahasa-cased` is a Malay model originally trained by `mesolitica`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_finetune_translation_super_tiny_standard_bahasa_cased_ms_4.3.0_3.0_1675102214685.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_finetune_translation_super_tiny_standard_bahasa_cased_ms_4.3.0_3.0_1675102214685.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_finetune_translation_super_tiny_standard_bahasa_cased","ms") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_finetune_translation_super_tiny_standard_bahasa_cased","ms") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_finetune_translation_super_tiny_standard_bahasa_cased| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|ms| +|Size:|37.7 MB| + +## References + +- https://huggingface.co/mesolitica/finetune-translation-t5-super-super-tiny-standard-bahasa-cased +- https://github.com/huseinzol05/malay-dataset/tree/master/translation/laser +- https://github.com/huseinzol05/malaya/tree/master/session/translation/hf-t5 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_finetune_translation_tiny_standard_bahasa_cased_ms.md b/docs/_posts/Cabir40/2023-01-30-t5_finetune_translation_tiny_standard_bahasa_cased_ms.md new file mode 100644 index 00000000000000..bb190471537f46 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_finetune_translation_tiny_standard_bahasa_cased_ms.md @@ -0,0 +1,86 @@ +--- +layout: model +title: Malay T5ForConditionalGeneration Tiny Cased model (from mesolitica) +author: John Snow Labs +name: t5_finetune_translation_tiny_standard_bahasa_cased +date: 2023-01-30 +tags: [ms, open_source, t5, tensorflow] +task: Text Generation +language: ms +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `finetune-translation-t5-tiny-standard-bahasa-cased` is a Malay model originally trained by `mesolitica`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_finetune_translation_tiny_standard_bahasa_cased_ms_4.3.0_3.0_1675102243431.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_finetune_translation_tiny_standard_bahasa_cased_ms_4.3.0_3.0_1675102243431.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_finetune_translation_tiny_standard_bahasa_cased","ms") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_finetune_translation_tiny_standard_bahasa_cased","ms") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_finetune_translation_tiny_standard_bahasa_cased| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|ms| +|Size:|176.9 MB| + +## References + +- https://huggingface.co/mesolitica/finetune-translation-t5-tiny-standard-bahasa-cased +- https://github.com/huseinzol05/malay-dataset/tree/master/translation/laser +- https://github.com/huseinzol05/malaya/tree/master/session/translation/hf-t5 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_flan_base_xx.md b/docs/_posts/Cabir40/2023-01-30-t5_flan_base_xx.md new file mode 100644 index 00000000000000..8be3f929c3654f --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_flan_base_xx.md @@ -0,0 +1,100 @@ +--- +layout: model +title: Multilingual T5ForConditionalGeneration Base Cased model (from google) +author: John Snow Labs +name: t5_flan_base +date: 2023-01-30 +tags: [vi, ne, fi, ur, ku, yo, si, ru, it, zh, la, hi, he, xh, so, ca, ar, as, sw, en, ro, ig, te, th, ta, ce, es, gu, or, fr, ka, "no", li, cr, ch, be, ha, ga, ja, pa, ko, sl, open_source, t5, xx, tensorflow] +task: Text Generation +language: xx +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `flan-t5-base` is a Multilingual model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_flan_base_xx_4.3.0_3.0_1675102308493.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_flan_base_xx_4.3.0_3.0_1675102308493.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_flan_base","xx") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_flan_base","xx") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_flan_base| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|xx| +|Size:|1.0 GB| + +## References + +- https://huggingface.co/google/flan-t5-base +- https://s3.amazonaws.com/moonup/production/uploads/1666363435475-62441d1d9fdefb55a0b7d12c.png +- https://github.com/google-research/t5x/blob/main/docs/models.md#flan-t5-checkpoints +- https://arxiv.org/pdf/2210.11416.pdf +- https://github.com/google-research/t5x +- https://arxiv.org/pdf/2210.11416.pdf +- https://arxiv.org/pdf/2210.11416.pdf +- https://arxiv.org/pdf/2210.11416.pdf +- https://s3.amazonaws.com/moonup/production/uploads/1666363265279-62441d1d9fdefb55a0b7d12c.png +- https://arxiv.org/pdf/2210.11416.pdf +- https://github.com/google-research/t5x +- https://github.com/google/jax +- https://s3.amazonaws.com/moonup/production/uploads/1668072995230-62441d1d9fdefb55a0b7d12c.png +- https://arxiv.org/pdf/2210.11416.pdf +- https://arxiv.org/pdf/2210.11416.pdf +- https://mlco2.github.io/impact#compute +- https://arxiv.org/abs/1910.09700 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_flan_small_xx.md b/docs/_posts/Cabir40/2023-01-30-t5_flan_small_xx.md new file mode 100644 index 00000000000000..e35b84877772f1 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_flan_small_xx.md @@ -0,0 +1,100 @@ +--- +layout: model +title: Multilingual T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_flan_small +date: 2023-01-30 +tags: [vi, ne, fi, ur, ku, yo, si, ru, it, zh, la, hi, he, xh, so, ca, ar, as, sw, en, ro, ig, te, th, ta, ce, es, gu, or, fr, ka, "no", li, cr, ch, be, ha, ga, ja, pa, ko, sl, open_source, t5, xx, tensorflow] +task: Text Generation +language: xx +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `flan-t5-small` is a Multilingual model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_flan_small_xx_4.3.0_3.0_1675102370004.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_flan_small_xx_4.3.0_3.0_1675102370004.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_flan_small","xx") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_flan_small","xx") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_flan_small| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|xx| +|Size:|349.5 MB| + +## References + +- https://huggingface.co/google/flan-t5-small +- https://s3.amazonaws.com/moonup/production/uploads/1666363435475-62441d1d9fdefb55a0b7d12c.png +- https://github.com/google-research/t5x/blob/main/docs/models.md#flan-t5-checkpoints +- https://arxiv.org/pdf/2210.11416.pdf +- https://github.com/google-research/t5x +- https://arxiv.org/pdf/2210.11416.pdf +- https://arxiv.org/pdf/2210.11416.pdf +- https://arxiv.org/pdf/2210.11416.pdf +- https://s3.amazonaws.com/moonup/production/uploads/1666363265279-62441d1d9fdefb55a0b7d12c.png +- https://arxiv.org/pdf/2210.11416.pdf +- https://github.com/google-research/t5x +- https://github.com/google/jax +- https://s3.amazonaws.com/moonup/production/uploads/1668072995230-62441d1d9fdefb55a0b7d12c.png +- https://arxiv.org/pdf/2210.11416.pdf +- https://arxiv.org/pdf/2210.11416.pdf +- https://mlco2.github.io/impact#compute +- https://arxiv.org/abs/1910.09700 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_friendly_ja.md b/docs/_posts/Cabir40/2023-01-30-t5_friendly_ja.md new file mode 100644 index 00000000000000..84bd59767dd31a --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_friendly_ja.md @@ -0,0 +1,87 @@ +--- +layout: model +title: Japanese T5ForConditionalGeneration Cased model (from astremo) +author: John Snow Labs +name: t5_friendly +date: 2023-01-30 +tags: [ja, open_source, t5, tensorflow] +task: Text Generation +language: ja +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `friendly_JA` is a Japanese model originally trained by `astremo`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_friendly_ja_4.3.0_3.0_1675102435483.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_friendly_ja_4.3.0_3.0_1675102435483.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_friendly","ja") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_friendly","ja") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_friendly| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|ja| +|Size:|923.1 MB| + +## References + +- https://huggingface.co/astremo/friendly_JA +- http://creativecommons.org/licenses/by/4.0/ +- http://creativecommons.org/licenses/by/4.0/ +- http://creativecommons.org/licenses/by/4.0/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_gagan3012_k2t_test_en.md b/docs/_posts/Cabir40/2023-01-30-t5_gagan3012_k2t_test_en.md new file mode 100644 index 00000000000000..b4a14468b02fd9 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_gagan3012_k2t_test_en.md @@ -0,0 +1,93 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from gagan3012) +author: John Snow Labs +name: t5_gagan3012_k2t_test +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `k2t-test` is a English model originally trained by `gagan3012`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_gagan3012_k2t_test_en_4.3.0_3.0_1675103953989.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_gagan3012_k2t_test_en_4.3.0_3.0_1675103953989.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_gagan3012_k2t_test","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_gagan3012_k2t_test","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_gagan3012_k2t_test| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|242.7 MB| + +## References + +- https://huggingface.co/gagan3012/k2t-test +- https://pypi.org/project/keytotext/ +- https://pepy.tech/project/keytotext +- https://colab.research.google.com/github/gagan3012/keytotext/blob/master/notebooks/K2T.ipynb +- https://share.streamlit.io/gagan3012/keytotext/UI/app.py +- https://github.com/gagan3012/keytotext#api +- https://hub.docker.com/r/gagan30/keytotext +- https://keytotext.readthedocs.io/en/latest/?badge=latest +- https://github.com/psf/black +- https://socialify.git.ci/gagan3012/keytotext/image?description=1&forks=1&language=1&owner=1&stargazers=1&theme=Light \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_gec_small_en.md b/docs/_posts/Cabir40/2023-01-30-t5_gec_small_en.md new file mode 100644 index 00000000000000..2cf50b50f308d4 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_gec_small_en.md @@ -0,0 +1,85 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from Unbabel) +author: John Snow Labs +name: t5_gec_small +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `gec-t5_small` is a English model originally trained by `Unbabel`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_gec_small_en_4.3.0_3.0_1675102492460.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_gec_small_en_4.3.0_3.0_1675102492460.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_gec_small","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_gec_small","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_gec_small| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|287.5 MB| + +## References + +- https://huggingface.co/Unbabel/gec-t5_small +- https://arxiv.org/pdf/2106.03830.pdf \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_gemini_small_en.md b/docs/_posts/Cabir40/2023-01-30-t5_gemini_small_en.md new file mode 100644 index 00000000000000..ca3c0f24b4725d --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_gemini_small_en.md @@ -0,0 +1,85 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from describeai) +author: John Snow Labs +name: t5_gemini_small +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `gemini-small` is a English model originally trained by `describeai`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_gemini_small_en_4.3.0_3.0_1675102559187.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_gemini_small_en_4.3.0_3.0_1675102559187.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_gemini_small","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_gemini_small","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_gemini_small| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|916.0 MB| + +## References + +- https://huggingface.co/describeai/gemini-small +- https://www.describe-ai.com/gemini \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_german_qg_e2e_quad_de.md b/docs/_posts/Cabir40/2023-01-30-t5_german_qg_e2e_quad_de.md new file mode 100644 index 00000000000000..918a91a5b63412 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_german_qg_e2e_quad_de.md @@ -0,0 +1,84 @@ +--- +layout: model +title: German T5ForConditionalGeneration Cased model (from dehio) +author: John Snow Labs +name: t5_german_qg_e2e_quad +date: 2023-01-30 +tags: [de, open_source, t5, tensorflow] +task: Text Generation +language: de +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `german-qg-t5-e2e-quad` is a German model originally trained by `dehio`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_german_qg_e2e_quad_de_4.3.0_3.0_1675102645662.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_german_qg_e2e_quad_de_4.3.0_3.0_1675102645662.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_german_qg_e2e_quad","de") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_german_qg_e2e_quad","de") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_german_qg_e2e_quad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|de| +|Size:|924.3 MB| + +## References + +- https://huggingface.co/dehio/german-qg-t5-e2e-quad \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_german_qg_quad_de.md b/docs/_posts/Cabir40/2023-01-30-t5_german_qg_quad_de.md new file mode 100644 index 00000000000000..6d2c4e3e248a08 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_german_qg_quad_de.md @@ -0,0 +1,86 @@ +--- +layout: model +title: German T5ForConditionalGeneration Cased model (from dehio) +author: John Snow Labs +name: t5_german_qg_quad +date: 2023-01-30 +tags: [de, open_source, t5, tensorflow] +task: Text Generation +language: de +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `german-qg-t5-quad` is a German model originally trained by `dehio`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_german_qg_quad_de_4.3.0_3.0_1675102735996.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_german_qg_quad_de_4.3.0_3.0_1675102735996.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_german_qg_quad","de") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_german_qg_quad","de") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_german_qg_quad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|de| +|Size:|923.3 MB| + +## References + +- https://huggingface.co/dehio/german-qg-t5-quad +- https://www.deepset.ai/germanquad +- https://github.com/d-e-h-i-o/german-qg \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_grammatical_error_correction_en.md b/docs/_posts/Cabir40/2023-01-30-t5_grammatical_error_correction_en.md new file mode 100644 index 00000000000000..551861bbf88736 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_grammatical_error_correction_en.md @@ -0,0 +1,84 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from leslyarun) +author: John Snow Labs +name: t5_grammatical_error_correction +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `grammatical-error-correction` is a English model originally trained by `leslyarun`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_grammatical_error_correction_en_4.3.0_3.0_1675102821479.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_grammatical_error_correction_en_4.3.0_3.0_1675102821479.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_grammatical_error_correction","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_grammatical_error_correction","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_grammatical_error_correction| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|920.7 MB| + +## References + +- https://huggingface.co/leslyarun/grammatical-error-correction \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_hupd_small_en.md b/docs/_posts/Cabir40/2023-01-30-t5_hupd_small_en.md new file mode 100644 index 00000000000000..e442218a8a226b --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_hupd_small_en.md @@ -0,0 +1,86 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from HUPD) +author: John Snow Labs +name: t5_hupd_small +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `hupd-t5-small` is a English model originally trained by `HUPD`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_hupd_small_en_4.3.0_3.0_1675102875669.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_hupd_small_en_4.3.0_3.0_1675102875669.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_hupd_small","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_hupd_small","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_hupd_small| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|289.1 MB| + +## References + +- https://huggingface.co/HUPD/hupd-t5-small +- https://patentdataset.org/ +- https://github.com/suzgunmirac/hupd \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_hybrid_hbh_small_ami_sum_en.md b/docs/_posts/Cabir40/2023-01-30-t5_hybrid_hbh_small_ami_sum_en.md new file mode 100644 index 00000000000000..a7992a7660bd33 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_hybrid_hbh_small_ami_sum_en.md @@ -0,0 +1,85 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from rohitsroch) +author: John Snow Labs +name: t5_hybrid_hbh_small_ami_sum +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `hybrid_hbh_t5-small_ami_sum` is a English model originally trained by `rohitsroch`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_hybrid_hbh_small_ami_sum_en_4.3.0_3.0_1675102906707.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_hybrid_hbh_small_ami_sum_en_4.3.0_3.0_1675102906707.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_hybrid_hbh_small_ami_sum","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_hybrid_hbh_small_ami_sum","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_hybrid_hbh_small_ami_sum| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|288.9 MB| + +## References + +- https://huggingface.co/rohitsroch/hybrid_hbh_t5-small_ami_sum +- https://doi.org/10.1145/3508546.3508640 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_idt5_qa_qg_id.md b/docs/_posts/Cabir40/2023-01-30-t5_idt5_qa_qg_id.md new file mode 100644 index 00000000000000..c7597801391852 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_idt5_qa_qg_id.md @@ -0,0 +1,89 @@ +--- +layout: model +title: Indonesian T5ForConditionalGeneration Cased model (from muchad) +author: John Snow Labs +name: t5_idt5_qa_qg +date: 2023-01-30 +tags: [id, open_source, t5, tensorflow] +task: Text Generation +language: id +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `idt5-qa-qg` is a Indonesian model originally trained by `muchad`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_idt5_qa_qg_id_4.3.0_3.0_1675102982556.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_idt5_qa_qg_id_4.3.0_3.0_1675102982556.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_idt5_qa_qg","id") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_idt5_qa_qg","id") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_idt5_qa_qg| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|id| +|Size:|981.3 MB| + +## References + +- https://huggingface.co/muchad/idt5-qa-qg +- https://github.com/Wikidepia/indonesian_datasets/tree/master/question-answering/squad +- https://ai.muchad.com/qg/ +- https://t.me/caritahubot +- https://colab.research.google.com/github/muchad/qaqg/blob/main/idT5_Question_Generation.ipynb +- https://colab.research.google.com/github/muchad/qaqg/blob/main/idT5_Question_Answering.ipynb \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_indot5_base_paraphrase_id.md b/docs/_posts/Cabir40/2023-01-30-t5_indot5_base_paraphrase_id.md new file mode 100644 index 00000000000000..fbcc3f79051ebe --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_indot5_base_paraphrase_id.md @@ -0,0 +1,83 @@ +--- +layout: model +title: Indonesian T5ForConditionalGeneration Base Cased model (from Wikidepia) +author: John Snow Labs +name: t5_indot5_base_paraphrase +date: 2023-01-30 +tags: [id, open_source, t5] +task: Text Generation +language: id +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `IndoT5-base-paraphrase` is a Indonesian model originally trained by `Wikidepia`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_indot5_base_paraphrase_id_4.3.0_3.0_1675097776595.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_indot5_base_paraphrase_id_4.3.0_3.0_1675097776595.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_indot5_base_paraphrase","id") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_indot5_base_paraphrase","id") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_indot5_base_paraphrase| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|id| +|Size:|1.0 GB| + +## References + +- https://huggingface.co/Wikidepia/IndoT5-base-paraphrase \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_indot5_small_id.md b/docs/_posts/Cabir40/2023-01-30-t5_indot5_small_id.md new file mode 100644 index 00000000000000..7192f390aba376 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_indot5_small_id.md @@ -0,0 +1,85 @@ +--- +layout: model +title: Indonesian T5ForConditionalGeneration Small Cased model (from Wikidepia) +author: John Snow Labs +name: t5_indot5_small +date: 2023-01-30 +tags: [id, open_source, t5, tensorflow] +task: Text Generation +language: id +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `IndoT5-small` is a Indonesian model originally trained by `Wikidepia`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_indot5_small_id_4.3.0_3.0_1675097879795.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_indot5_small_id_4.3.0_3.0_1675097879795.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_indot5_small","id") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_indot5_small","id") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_indot5_small| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|id| +|Size:|179.1 MB| + +## References + +- https://huggingface.co/Wikidepia/IndoT5-small +- https://github.com/Wikidepia/indonesian_datasets/tree/master/dump/mc4 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_informal_formal_style_transfer_en.md b/docs/_posts/Cabir40/2023-01-30-t5_informal_formal_style_transfer_en.md new file mode 100644 index 00000000000000..a63b9cc27138d1 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_informal_formal_style_transfer_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from rajistics) +author: John Snow Labs +name: t5_informal_formal_style_transfer +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `informal_formal_style_transfer` is a English model originally trained by `rajistics`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_informal_formal_style_transfer_en_4.3.0_3.0_1675103071459.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_informal_formal_style_transfer_en_4.3.0_3.0_1675103071459.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_informal_formal_style_transfer","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_informal_formal_style_transfer","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_informal_formal_style_transfer| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|924.5 MB| + +## References + +- https://huggingface.co/rajistics/informal_formal_style_transfer +- https://github.com/PrithivirajDamodaran/Styleformer +- https://www.aclweb.org/anthology/D19-5502.pdf +- http://cs230.stanford.edu/projects_winter_2020/reports/32069807.pdf +- https://arxiv.org/pdf/1804.06437.pdf \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_it5_base_news_summarization_it.md b/docs/_posts/Cabir40/2023-01-30-t5_it5_base_news_summarization_it.md new file mode 100644 index 00000000000000..90e27279919a79 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_it5_base_news_summarization_it.md @@ -0,0 +1,89 @@ +--- +layout: model +title: Italian T5ForConditionalGeneration Base Cased model (from it5) +author: John Snow Labs +name: t5_it5_base_news_summarization +date: 2023-01-30 +tags: [it, open_source, t5, tensorflow] +task: Text Generation +language: it +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `it5-base-news-summarization` is a Italian model originally trained by `it5`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_it5_base_news_summarization_it_4.3.0_3.0_1675103162283.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_it5_base_news_summarization_it_4.3.0_3.0_1675103162283.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_it5_base_news_summarization","it") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_it5_base_news_summarization","it") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_it5_base_news_summarization| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|it| +|Size:|1.0 GB| + +## References + +- https://huggingface.co/it5/it5-base-news-summarization +- https://arxiv.org/abs/2203.03759 +- https://gsarti.com +- https://malvinanissim.github.io +- https://github.com/gsarti/it5 +- https://paperswithcode.com/sota?task=News+Summarization&dataset=NewsSum-IT \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_it5_efficient_small_el32_formal_to_informal_it.md b/docs/_posts/Cabir40/2023-01-30-t5_it5_efficient_small_el32_formal_to_informal_it.md new file mode 100644 index 00000000000000..5a1c744a4565a1 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_it5_efficient_small_el32_formal_to_informal_it.md @@ -0,0 +1,91 @@ +--- +layout: model +title: Italian T5ForConditionalGeneration Small Cased model (from it5) +author: John Snow Labs +name: t5_it5_efficient_small_el32_formal_to_informal +date: 2023-01-30 +tags: [it, open_source, t5, tensorflow] +task: Text Generation +language: it +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `it5-efficient-small-el32-formal-to-informal` is a Italian model originally trained by `it5`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_it5_efficient_small_el32_formal_to_informal_it_4.3.0_3.0_1675103238992.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_it5_efficient_small_el32_formal_to_informal_it_4.3.0_3.0_1675103238992.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_it5_efficient_small_el32_formal_to_informal","it") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_it5_efficient_small_el32_formal_to_informal","it") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_it5_efficient_small_el32_formal_to_informal| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|it| +|Size:|593.5 MB| + +## References + +- https://huggingface.co/it5/it5-efficient-small-el32-formal-to-informal +- https://github.com/stefan-it +- https://arxiv.org/abs/2203.03759 +- https://gsarti.com +- https://malvinanissim.github.io +- https://arxiv.org/abs/2109.10686 +- https://github.com/gsarti/it5 +- https://paperswithcode.com/sota?task=Formal-to-informal+Style+Transfer&dataset=XFORMAL+%28Italian+Subset%29 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_it5_efficient_small_el32_headline_generation_it.md b/docs/_posts/Cabir40/2023-01-30-t5_it5_efficient_small_el32_headline_generation_it.md new file mode 100644 index 00000000000000..82da9c32154d6d --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_it5_efficient_small_el32_headline_generation_it.md @@ -0,0 +1,91 @@ +--- +layout: model +title: Italian T5ForConditionalGeneration Small Cased model (from it5) +author: John Snow Labs +name: t5_it5_efficient_small_el32_headline_generation +date: 2023-01-30 +tags: [it, open_source, t5, tensorflow] +task: Text Generation +language: it +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `it5-efficient-small-el32-headline-generation` is a Italian model originally trained by `it5`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_it5_efficient_small_el32_headline_generation_it_4.3.0_3.0_1675103295731.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_it5_efficient_small_el32_headline_generation_it_4.3.0_3.0_1675103295731.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_it5_efficient_small_el32_headline_generation","it") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_it5_efficient_small_el32_headline_generation","it") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_it5_efficient_small_el32_headline_generation| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|it| +|Size:|594.0 MB| + +## References + +- https://huggingface.co/it5/it5-efficient-small-el32-headline-generation +- https://github.com/stefan-it +- https://arxiv.org/abs/2203.03759 +- https://gsarti.com +- https://malvinanissim.github.io +- https://arxiv.org/abs/2109.10686 +- https://github.com/gsarti/it5 +- https://paperswithcode.com/sota?task=Headline+generation&dataset=HeadGen-IT \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_it5_efficient_small_el32_ilgiornale_to_repubblica_it.md b/docs/_posts/Cabir40/2023-01-30-t5_it5_efficient_small_el32_ilgiornale_to_repubblica_it.md new file mode 100644 index 00000000000000..e9c1f0478debd5 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_it5_efficient_small_el32_ilgiornale_to_repubblica_it.md @@ -0,0 +1,91 @@ +--- +layout: model +title: Italian T5ForConditionalGeneration Small Cased model (from it5) +author: John Snow Labs +name: t5_it5_efficient_small_el32_ilgiornale_to_repubblica +date: 2023-01-30 +tags: [it, open_source, t5, tensorflow] +task: Text Generation +language: it +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `it5-efficient-small-el32-ilgiornale-to-repubblica` is a Italian model originally trained by `it5`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_it5_efficient_small_el32_ilgiornale_to_repubblica_it_4.3.0_3.0_1675103353037.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_it5_efficient_small_el32_ilgiornale_to_repubblica_it_4.3.0_3.0_1675103353037.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_it5_efficient_small_el32_ilgiornale_to_repubblica","it") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_it5_efficient_small_el32_ilgiornale_to_repubblica","it") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_it5_efficient_small_el32_ilgiornale_to_repubblica| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|it| +|Size:|593.9 MB| + +## References + +- https://huggingface.co/it5/it5-efficient-small-el32-ilgiornale-to-repubblica +- https://github.com/stefan-it +- https://arxiv.org/abs/2203.03759 +- https://gsarti.com +- https://malvinanissim.github.io +- https://arxiv.org/abs/2109.10686 +- https://github.com/gsarti/it5 +- https://paperswithcode.com/sota?task=Headline+style+transfer+%28Il+Giornale+to+Repubblica%29&dataset=CHANGE-IT \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_it5_efficient_small_el32_informal_to_formal_it.md b/docs/_posts/Cabir40/2023-01-30-t5_it5_efficient_small_el32_informal_to_formal_it.md new file mode 100644 index 00000000000000..9860f2b68ecfa9 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_it5_efficient_small_el32_informal_to_formal_it.md @@ -0,0 +1,91 @@ +--- +layout: model +title: Italian T5ForConditionalGeneration Small Cased model (from it5) +author: John Snow Labs +name: t5_it5_efficient_small_el32_informal_to_formal +date: 2023-01-30 +tags: [it, open_source, t5, tensorflow] +task: Text Generation +language: it +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `it5-efficient-small-el32-informal-to-formal` is a Italian model originally trained by `it5`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_it5_efficient_small_el32_informal_to_formal_it_4.3.0_3.0_1675103414416.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_it5_efficient_small_el32_informal_to_formal_it_4.3.0_3.0_1675103414416.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_it5_efficient_small_el32_informal_to_formal","it") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_it5_efficient_small_el32_informal_to_formal","it") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_it5_efficient_small_el32_informal_to_formal| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|it| +|Size:|593.5 MB| + +## References + +- https://huggingface.co/it5/it5-efficient-small-el32-informal-to-formal +- https://github.com/stefan-it +- https://arxiv.org/abs/2203.03759 +- https://gsarti.com +- https://malvinanissim.github.io +- https://arxiv.org/abs/2109.10686 +- https://github.com/gsarti/it5 +- https://paperswithcode.com/sota?task=Informal-to-formal+Style+Transfer&dataset=XFORMAL+%28Italian+Subset%29 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_it5_efficient_small_el32_news_summarization_it.md b/docs/_posts/Cabir40/2023-01-30-t5_it5_efficient_small_el32_news_summarization_it.md new file mode 100644 index 00000000000000..cc5713fce9dc07 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_it5_efficient_small_el32_news_summarization_it.md @@ -0,0 +1,91 @@ +--- +layout: model +title: Italian T5ForConditionalGeneration Small Cased model (from it5) +author: John Snow Labs +name: t5_it5_efficient_small_el32_news_summarization +date: 2023-01-30 +tags: [it, open_source, t5, tensorflow] +task: Text Generation +language: it +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `it5-efficient-small-el32-news-summarization` is a Italian model originally trained by `it5`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_it5_efficient_small_el32_news_summarization_it_4.3.0_3.0_1675103478274.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_it5_efficient_small_el32_news_summarization_it_4.3.0_3.0_1675103478274.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_it5_efficient_small_el32_news_summarization","it") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_it5_efficient_small_el32_news_summarization","it") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_it5_efficient_small_el32_news_summarization| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|it| +|Size:|594.0 MB| + +## References + +- https://huggingface.co/it5/it5-efficient-small-el32-news-summarization +- https://github.com/stefan-it +- https://arxiv.org/abs/2203.03759 +- https://gsarti.com +- https://malvinanissim.github.io +- https://arxiv.org/abs/2109.10686 +- https://github.com/gsarti/it5 +- https://paperswithcode.com/sota?task=News+Summarization&dataset=NewsSum-IT \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_it5_efficient_small_el32_question_answering_it.md b/docs/_posts/Cabir40/2023-01-30-t5_it5_efficient_small_el32_question_answering_it.md new file mode 100644 index 00000000000000..5612d2ea75a0a5 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_it5_efficient_small_el32_question_answering_it.md @@ -0,0 +1,90 @@ +--- +layout: model +title: Italian T5ForConditionalGeneration Small Cased model (from it5) +author: John Snow Labs +name: t5_it5_efficient_small_el32_question_answering +date: 2023-01-30 +tags: [it, open_source, t5, tensorflow] +task: Text Generation +language: it +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `it5-efficient-small-el32-question-answering` is a Italian model originally trained by `it5`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_it5_efficient_small_el32_question_answering_it_4.3.0_3.0_1675103534190.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_it5_efficient_small_el32_question_answering_it_4.3.0_3.0_1675103534190.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_it5_efficient_small_el32_question_answering","it") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_it5_efficient_small_el32_question_answering","it") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_it5_efficient_small_el32_question_answering| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|it| +|Size:|593.9 MB| + +## References + +- https://huggingface.co/it5/it5-efficient-small-el32-question-answering +- https://github.com/stefan-it +- https://arxiv.org/abs/2203.03759 +- https://gsarti.com +- https://malvinanissim.github.io +- https://github.com/gsarti/it5 +- https://paperswithcode.com/sota?task=Question+Answering&dataset=SQuAD-IT \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_it5_efficient_small_el32_question_generation_it.md b/docs/_posts/Cabir40/2023-01-30-t5_it5_efficient_small_el32_question_generation_it.md new file mode 100644 index 00000000000000..316fb9168e212c --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_it5_efficient_small_el32_question_generation_it.md @@ -0,0 +1,91 @@ +--- +layout: model +title: Italian T5ForConditionalGeneration Small Cased model (from it5) +author: John Snow Labs +name: t5_it5_efficient_small_el32_question_generation +date: 2023-01-30 +tags: [it, open_source, t5, tensorflow] +task: Text Generation +language: it +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `it5-efficient-small-el32-question-generation` is a Italian model originally trained by `it5`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_it5_efficient_small_el32_question_generation_it_4.3.0_3.0_1675103595829.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_it5_efficient_small_el32_question_generation_it_4.3.0_3.0_1675103595829.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_it5_efficient_small_el32_question_generation","it") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_it5_efficient_small_el32_question_generation","it") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_it5_efficient_small_el32_question_generation| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|it| +|Size:|593.8 MB| + +## References + +- https://huggingface.co/it5/it5-efficient-small-el32-question-generation +- https://github.com/stefan-it +- https://arxiv.org/abs/2203.03759 +- https://gsarti.com +- https://malvinanissim.github.io +- https://arxiv.org/abs/2109.10686 +- https://github.com/gsarti/it5 +- https://paperswithcode.com/sota?task=Question+generation&dataset=SQuAD-IT \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_it5_efficient_small_el32_repubblica_to_ilgiornale_it.md b/docs/_posts/Cabir40/2023-01-30-t5_it5_efficient_small_el32_repubblica_to_ilgiornale_it.md new file mode 100644 index 00000000000000..684063bde21eaa --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_it5_efficient_small_el32_repubblica_to_ilgiornale_it.md @@ -0,0 +1,91 @@ +--- +layout: model +title: Italian T5ForConditionalGeneration Small Cased model (from it5) +author: John Snow Labs +name: t5_it5_efficient_small_el32_repubblica_to_ilgiornale +date: 2023-01-30 +tags: [it, open_source, t5, tensorflow] +task: Text Generation +language: it +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `it5-efficient-small-el32-repubblica-to-ilgiornale` is a Italian model originally trained by `it5`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_it5_efficient_small_el32_repubblica_to_ilgiornale_it_4.3.0_3.0_1675103650043.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_it5_efficient_small_el32_repubblica_to_ilgiornale_it_4.3.0_3.0_1675103650043.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_it5_efficient_small_el32_repubblica_to_ilgiornale","it") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_it5_efficient_small_el32_repubblica_to_ilgiornale","it") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_it5_efficient_small_el32_repubblica_to_ilgiornale| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|it| +|Size:|594.0 MB| + +## References + +- https://huggingface.co/it5/it5-efficient-small-el32-repubblica-to-ilgiornale +- https://github.com/stefan-it +- https://arxiv.org/abs/2203.03759 +- https://gsarti.com +- https://malvinanissim.github.io +- https://arxiv.org/abs/2109.10686 +- https://github.com/gsarti/it5 +- https://paperswithcode.com/sota?task=Headline+style+transfer+%28Repubblica+to+Il+Giornale%29&dataset=CHANGE-IT \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_it5_efficient_small_el32_wiki_summarization_it.md b/docs/_posts/Cabir40/2023-01-30-t5_it5_efficient_small_el32_wiki_summarization_it.md new file mode 100644 index 00000000000000..8292efe76750d6 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_it5_efficient_small_el32_wiki_summarization_it.md @@ -0,0 +1,92 @@ +--- +layout: model +title: Italian T5ForConditionalGeneration Small Cased model (from it5) +author: John Snow Labs +name: t5_it5_efficient_small_el32_wiki_summarization +date: 2023-01-30 +tags: [it, open_source, t5, tensorflow] +task: Text Generation +language: it +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `it5-efficient-small-el32-wiki-summarization` is a Italian model originally trained by `it5`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_it5_efficient_small_el32_wiki_summarization_it_4.3.0_3.0_1675103710830.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_it5_efficient_small_el32_wiki_summarization_it_4.3.0_3.0_1675103710830.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_it5_efficient_small_el32_wiki_summarization","it") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_it5_efficient_small_el32_wiki_summarization","it") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_it5_efficient_small_el32_wiki_summarization| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|it| +|Size:|594.0 MB| + +## References + +- https://huggingface.co/it5/it5-efficient-small-el32-wiki-summarization +- https://github.com/stefan-it +- https://www.semanticscholar.org/paper/WITS%3A-Wikipedia-for-Italian-Text-Summarization-Casola-Lavelli/ad6c83122e721c7c0db4a40727dac3b4762cd2b1 +- https://arxiv.org/abs/2203.03759 +- https://gsarti.com +- https://malvinanissim.github.io +- https://arxiv.org/abs/2109.10686 +- https://github.com/gsarti/it5 +- https://paperswithcode.com/sota?task=Wikipedia+Summarization&dataset=WITS \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_it5_efficient_small_fanpage_it.md b/docs/_posts/Cabir40/2023-01-30-t5_it5_efficient_small_fanpage_it.md new file mode 100644 index 00000000000000..79dd7a1264417f --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_it5_efficient_small_fanpage_it.md @@ -0,0 +1,84 @@ +--- +layout: model +title: Italian T5ForConditionalGeneration Small Cased model (from efederici) +author: John Snow Labs +name: t5_it5_efficient_small_fanpage +date: 2023-01-30 +tags: [it, open_source, t5, tensorflow] +task: Text Generation +language: it +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `it5-efficient-small-fanpage` is a Italian model originally trained by `efederici`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_it5_efficient_small_fanpage_it_4.3.0_3.0_1675103771407.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_it5_efficient_small_fanpage_it_4.3.0_3.0_1675103771407.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_it5_efficient_small_fanpage","it") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_it5_efficient_small_fanpage","it") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_it5_efficient_small_fanpage| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|it| +|Size:|593.8 MB| + +## References + +- https://huggingface.co/efederici/it5-efficient-small-fanpage \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_it5_efficient_small_lfqa_it.md b/docs/_posts/Cabir40/2023-01-30-t5_it5_efficient_small_lfqa_it.md new file mode 100644 index 00000000000000..466cd2c9d6dd11 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_it5_efficient_small_lfqa_it.md @@ -0,0 +1,84 @@ +--- +layout: model +title: Italian T5ForConditionalGeneration Small Cased model (from efederici) +author: John Snow Labs +name: t5_it5_efficient_small_lfqa +date: 2023-01-30 +tags: [it, open_source, t5, tensorflow] +task: Text Generation +language: it +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `it5-efficient-small-lfqa` is a Italian model originally trained by `efederici`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_it5_efficient_small_lfqa_it_4.3.0_3.0_1675103827826.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_it5_efficient_small_lfqa_it_4.3.0_3.0_1675103827826.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_it5_efficient_small_lfqa","it") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_it5_efficient_small_lfqa","it") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_it5_efficient_small_lfqa| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|it| +|Size:|594.0 MB| + +## References + +- https://huggingface.co/efederici/it5-efficient-small-lfqa \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_jainu_ja.md b/docs/_posts/Cabir40/2023-01-30-t5_jainu_ja.md new file mode 100644 index 00000000000000..3ca3baa8ad5b03 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_jainu_ja.md @@ -0,0 +1,86 @@ +--- +layout: model +title: Japanese T5ForConditionalGeneration Cased model (from astremo) +author: John Snow Labs +name: t5_jainu +date: 2023-01-30 +tags: [ja, open_source, t5] +task: Text Generation +language: ja +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `JAINU` is a Japanese model originally trained by `astremo`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_jainu_ja_4.3.0_3.0_1675097938002.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_jainu_ja_4.3.0_3.0_1675097938002.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_jainu","ja") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_jainu","ja") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_jainu| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|ja| +|Size:|923.5 MB| + +## References + +- https://huggingface.co/astremo/JAINU +- http://creativecommons.org/licenses/by/4.0/ +- http://creativecommons.org/licenses/by/4.0/ +- http://creativecommons.org/licenses/by/4.0/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_k2t_new_en.md b/docs/_posts/Cabir40/2023-01-30-t5_k2t_new_en.md new file mode 100644 index 00000000000000..431318ff918efe --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_k2t_new_en.md @@ -0,0 +1,96 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from gagan3012) +author: John Snow Labs +name: t5_k2t_new +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `k2t-new` is a English model originally trained by `gagan3012`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_k2t_new_en_4.3.0_3.0_1675103876567.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_k2t_new_en_4.3.0_3.0_1675103876567.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_k2t_new","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_k2t_new","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_k2t_new| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|262.7 MB| + +## References + +- https://huggingface.co/gagan3012/k2t-new +- https://user-images.githubusercontent.com/49101362/116334480-f5e57a00-a7dd-11eb-987c-186477f94b6e.png +- https://pypi.org/project/keytotext/ +- https://pepy.tech/project/keytotext +- https://colab.research.google.com/github/gagan3012/keytotext/blob/master/Examples/K2T.ipynb +- https://share.streamlit.io/gagan3012/keytotext/UI/app.py +- https://github.com/gagan3012/keytotext/tree/master/Training%20Notebooks +- https://colab.research.google.com/github/gagan3012/keytotext/blob/master/Examples/K2T.ipynb +- https://github.com/gagan3012/keytotext/tree/master/Examples +- https://user-images.githubusercontent.com/49101362/116220679-90e64180-a755-11eb-9246-82d93d924a6c.png +- https://share.streamlit.io/gagan3012/keytotext/UI/app.py +- https://github.com/gagan3012/streamlit-tags +- https://user-images.githubusercontent.com/49101362/116162205-fc042980-a6fd-11eb-892e-8f6902f193f4.png \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_k2t_test3_en.md b/docs/_posts/Cabir40/2023-01-30-t5_k2t_test3_en.md new file mode 100644 index 00000000000000..86b1f2f0fcfa25 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_k2t_test3_en.md @@ -0,0 +1,93 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from gagan3012) +author: John Snow Labs +name: t5_k2t_test3 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `k2t-test3` is a English model originally trained by `gagan3012`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_k2t_test3_en_4.3.0_3.0_1675103993995.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_k2t_test3_en_4.3.0_3.0_1675103993995.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_k2t_test3","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_k2t_test3","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_k2t_test3| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|246.4 MB| + +## References + +- https://huggingface.co/gagan3012/k2t-test3 +- https://pypi.org/project/keytotext/ +- https://pepy.tech/project/keytotext +- https://colab.research.google.com/github/gagan3012/keytotext/blob/master/notebooks/K2T.ipynb +- https://share.streamlit.io/gagan3012/keytotext/UI/app.py +- https://github.com/gagan3012/keytotext#api +- https://hub.docker.com/r/gagan30/keytotext +- https://keytotext.readthedocs.io/en/latest/?badge=latest +- https://github.com/psf/black +- https://socialify.git.ci/gagan3012/keytotext/image?description=1&forks=1&language=1&owner=1&stargazers=1&theme=Light \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_kbd_lat_835k_3m_small_ru.md b/docs/_posts/Cabir40/2023-01-30-t5_kbd_lat_835k_3m_small_ru.md new file mode 100644 index 00000000000000..24f58a69849a9d --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_kbd_lat_835k_3m_small_ru.md @@ -0,0 +1,84 @@ +--- +layout: model +title: Russian T5ForConditionalGeneration Small Cased model (from anzorq) +author: John Snow Labs +name: t5_kbd_lat_835k_3m_small +date: 2023-01-30 +tags: [ru, open_source, t5, tensorflow] +task: Text Generation +language: ru +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `kbd_lat-835k_ru-3M_t5-small` is a Russian model originally trained by `anzorq`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_kbd_lat_835k_3m_small_ru_4.3.0_3.0_1675104021877.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_kbd_lat_835k_3m_small_ru_4.3.0_3.0_1675104021877.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_kbd_lat_835k_3m_small","ru") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_kbd_lat_835k_3m_small","ru") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_kbd_lat_835k_3m_small| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|ru| +|Size:|211.3 MB| + +## References + +- https://huggingface.co/anzorq/kbd_lat-835k_ru-3M_t5-small \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_kbd_lat_char_tokenizer_ru.md b/docs/_posts/Cabir40/2023-01-30-t5_kbd_lat_char_tokenizer_ru.md new file mode 100644 index 00000000000000..df7f950a0b5020 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_kbd_lat_char_tokenizer_ru.md @@ -0,0 +1,84 @@ +--- +layout: model +title: Russian T5ForConditionalGeneration Cased model (from anzorq) +author: John Snow Labs +name: t5_kbd_lat_char_tokenizer +date: 2023-01-30 +tags: [ru, open_source, t5, tensorflow] +task: Text Generation +language: ru +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `kbd_lat-ru_char_tokenizer` is a Russian model originally trained by `anzorq`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_kbd_lat_char_tokenizer_ru_4.3.0_3.0_1675104071281.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_kbd_lat_char_tokenizer_ru_4.3.0_3.0_1675104071281.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_kbd_lat_char_tokenizer","ru") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_kbd_lat_char_tokenizer","ru") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_kbd_lat_char_tokenizer| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|ru| +|Size:|777.2 MB| + +## References + +- https://huggingface.co/anzorq/kbd_lat-ru_char_tokenizer \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_ke_base_ko.md b/docs/_posts/Cabir40/2023-01-30-t5_ke_base_ko.md new file mode 100644 index 00000000000000..b9e63b0207fa56 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_ke_base_ko.md @@ -0,0 +1,95 @@ +--- +layout: model +title: Korean T5ForConditionalGeneration Base Cased model (from KETI-AIR) +author: John Snow Labs +name: t5_ke_base +date: 2023-01-30 +tags: [ko, open_source, t5, tensorflow] +task: Text Generation +language: ko +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `ke-t5-base-ko` is a Korean model originally trained by `KETI-AIR`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_ke_base_ko_4.3.0_3.0_1675104551769.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_ke_base_ko_4.3.0_3.0_1675104551769.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_ke_base","ko") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_ke_base","ko") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_ke_base| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|ko| +|Size:|569.1 MB| + +## References + +- https://huggingface.co/KETI-AIR/ke-t5-base-ko +- https://github.com/google-research/text-to-text-transfer-transformer#released-model-checkpoints +- https://github.com/AIRC-KETI/ke-t5 +- https://aclanthology.org/2021.findings-emnlp.33/ +- https://jmlr.org/papers/volume21/20-074/20-074.pdf +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://aclanthology.org/2021.acl-long.330.pdf +- https://dl.acm.org/doi/pdf/10.1145/3442188.3445922 +- https://www.tensorflow.org/datasets/catalog/c4 +- https://jmlr.org/papers/volume21/20-074/20-074.pdf +- https://mlco2.github.io/impact#compute +- https://arxiv.org/abs/1910.09700 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_ke_base_xx.md b/docs/_posts/Cabir40/2023-01-30-t5_ke_base_xx.md new file mode 100644 index 00000000000000..67a04e12aa291a --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_ke_base_xx.md @@ -0,0 +1,100 @@ +--- +layout: model +title: Multilingual T5ForConditionalGeneration Base Cased model (from KETI-AIR) +author: John Snow Labs +name: t5_ke_base +date: 2023-01-30 +tags: [en, ko, open_source, t5, xx, tensorflow] +task: Text Generation +language: xx +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `ke-t5-base` is a Multilingual model originally trained by `KETI-AIR`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_ke_base_xx_4.3.0_3.0_1675104312892.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_ke_base_xx_4.3.0_3.0_1675104312892.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_ke_base","xx") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_ke_base","xx") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_ke_base| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|xx| +|Size:|569.3 MB| + +## References + +- https://huggingface.co/KETI-AIR/ke-t5-base +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://github.com/google-research/text-to-text-transfer-transformer#released-model-checkpoints +- https://github.com/AIRC-KETI/ke-t5 +- https://aclanthology.org/2021.findings-emnlp.33/ +- https://jmlr.org/papers/volume21/20-074/20-074.pdf +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://aclanthology.org/2021.acl-long.330.pdf +- https://dl.acm.org/doi/pdf/10.1145/3442188.3445922 +- https://www.tensorflow.org/datasets/catalog/c4 +- https://jmlr.org/papers/volume21/20-074/20-074.pdf +- https://jmlr.org/papers/volume21/20-074/20-074.pdf +- https://jmlr.org/papers/volume21/20-074/20-074.pdf +- https://mlco2.github.io/impact#compute +- https://arxiv.org/abs/1910.09700 +- https://colab.research.google.com/github/google-research/text-to-text-transfer-transformer/blob/main/notebooks/t5-trivia.ipynb \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_ke_small_ko.md b/docs/_posts/Cabir40/2023-01-30-t5_ke_small_ko.md new file mode 100644 index 00000000000000..5afb43b78b4f0c --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_ke_small_ko.md @@ -0,0 +1,87 @@ +--- +layout: model +title: Korean T5ForConditionalGeneration Small Cased model (from KETI-AIR) +author: John Snow Labs +name: t5_ke_small +date: 2023-01-30 +tags: [ko, open_source, t5, tensorflow] +task: Text Generation +language: ko +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `ke-t5-small-ko` is a Korean model originally trained by `KETI-AIR`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_ke_small_ko_4.3.0_3.0_1675104654891.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_ke_small_ko_4.3.0_3.0_1675104654891.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_ke_small","ko") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_ke_small","ko") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_ke_small| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|ko| +|Size:|211.1 MB| + +## References + +- https://huggingface.co/KETI-AIR/ke-t5-small-ko +- https://github.com/AIRC-KETI/ke-t5 +- https://aclanthology.org/2021.findings-emnlp.33/ +- https://koreascience.kr/article/CFKO202130060717834.pdf \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_kes_en.md b/docs/_posts/Cabir40/2023-01-30-t5_kes_en.md new file mode 100644 index 00000000000000..bb8302763a18d8 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_kes_en.md @@ -0,0 +1,87 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from KES) +author: John Snow Labs +name: t5_kes +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `T5-KES` is a English model originally trained by `KES`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_kes_en_4.3.0_3.0_1675099343508.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_kes_en_4.3.0_3.0_1675099343508.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_kes","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_kes","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_kes| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|912.8 MB| + +## References + +- https://huggingface.co/KES/T5-KES +- https://arxiv.org/abs/1702.04066 +- https://github.com/EricFillion/happy-transformer +- https://pypi.org/project/Caribe/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_keyphrase_generation_small_inspec_en.md b/docs/_posts/Cabir40/2023-01-30-t5_keyphrase_generation_small_inspec_en.md new file mode 100644 index 00000000000000..911a9846983c9f --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_keyphrase_generation_small_inspec_en.md @@ -0,0 +1,86 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from ml6team) +author: John Snow Labs +name: t5_keyphrase_generation_small_inspec +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `keyphrase-generation-t5-small-inspec` is a English model originally trained by `ml6team`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_keyphrase_generation_small_inspec_en_4.3.0_3.0_1675104684365.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_keyphrase_generation_small_inspec_en_4.3.0_3.0_1675104684365.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_keyphrase_generation_small_inspec","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_keyphrase_generation_small_inspec","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_keyphrase_generation_small_inspec| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|280.5 MB| + +## References + +- https://huggingface.co/ml6team/keyphrase-generation-t5-small-inspec +- https://dl.acm.org/doi/10.3115/1119355.1119383 +- https://paperswithcode.com/sota?task=Keyphrase+Generation&dataset=inspec \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_keyphrase_generation_small_openkp_en.md b/docs/_posts/Cabir40/2023-01-30-t5_keyphrase_generation_small_openkp_en.md new file mode 100644 index 00000000000000..9f9ed7915d69f2 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_keyphrase_generation_small_openkp_en.md @@ -0,0 +1,87 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from ml6team) +author: John Snow Labs +name: t5_keyphrase_generation_small_openkp +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `keyphrase-generation-t5-small-openkp` is a English model originally trained by `ml6team`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_keyphrase_generation_small_openkp_en_4.3.0_3.0_1675104714518.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_keyphrase_generation_small_openkp_en_4.3.0_3.0_1675104714518.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_keyphrase_generation_small_openkp","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_keyphrase_generation_small_openkp","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_keyphrase_generation_small_openkp| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|285.9 MB| + +## References + +- https://huggingface.co/ml6team/keyphrase-generation-t5-small-openkp +- https://github.com/microsoft/OpenKP +- https://arxiv.org/abs/1911.02671 +- https://paperswithcode.com/sota?task=Keyphrase+Generation&dataset=openkp \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_keyt5_base_ru.md b/docs/_posts/Cabir40/2023-01-30-t5_keyt5_base_ru.md new file mode 100644 index 00000000000000..ef57449591ac69 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_keyt5_base_ru.md @@ -0,0 +1,90 @@ +--- +layout: model +title: Russian T5ForConditionalGeneration Base Cased model (from 0x7194633) +author: John Snow Labs +name: t5_keyt5_base +date: 2023-01-30 +tags: [ru, open_source, t5, tensorflow] +task: Text Generation +language: ru +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `keyt5-base` is a Russian model originally trained by `0x7194633`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_keyt5_base_ru_4.3.0_3.0_1675104774932.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_keyt5_base_ru_4.3.0_3.0_1675104774932.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_keyt5_base","ru") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_keyt5_base","ru") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_keyt5_base| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|ru| +|Size:|927.4 MB| + +## References + +- https://huggingface.co/0x7194633/keyt5-base +- https://github.com/0x7o/text2keywords +- https://github.com/0x7o/text2keywords +- https://github.com/0x7o/text2keywords +- https://github.com/0x7o/text2keywords +- https://colab.research.google.com/github/0x7o/text2keywords/blob/main/example/keyT5_use.ipynb +- https://colab.research.google.com/github/0x7o/text2keywords/blob/main/example/keyT5_train.ipynb \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_kingjamesify_base_en.md b/docs/_posts/Cabir40/2023-01-30-t5_kingjamesify_base_en.md new file mode 100644 index 00000000000000..1654b1cf840d56 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_kingjamesify_base_en.md @@ -0,0 +1,83 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from swcrazyfan) +author: John Snow Labs +name: t5_kingjamesify_base +date: 2023-01-30 +tags: [en, open_source, t5] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `KingJamesify-T5-Base` is a English model originally trained by `swcrazyfan`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_kingjamesify_base_en_4.3.0_3.0_1675098022215.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_kingjamesify_base_en_4.3.0_3.0_1675098022215.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_kingjamesify_base","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_kingjamesify_base","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_kingjamesify_base| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|915.1 MB| + +## References + +- https://huggingface.co/swcrazyfan/KingJamesify-T5-Base \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_legacy_sl_small_sl.md b/docs/_posts/Cabir40/2023-01-30-t5_legacy_sl_small_sl.md new file mode 100644 index 00000000000000..1f0edae22ac893 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_legacy_sl_small_sl.md @@ -0,0 +1,84 @@ +--- +layout: model +title: Slovenian T5ForConditionalGeneration Small Cased model (from cjvt) +author: John Snow Labs +name: t5_legacy_sl_small +date: 2023-01-30 +tags: [sl, open_source, t5, tensorflow] +task: Text Generation +language: sl +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `legacy-t5-sl-small` is a Slovenian model originally trained by `cjvt`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_legacy_sl_small_sl_4.3.0_3.0_1675104880094.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_legacy_sl_small_sl_4.3.0_3.0_1675104880094.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_legacy_sl_small","sl") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_legacy_sl_small","sl") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_legacy_sl_small| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|sl| +|Size:|178.9 MB| + +## References + +- https://huggingface.co/cjvt/legacy-t5-sl-small \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_lewip_informal_en.md b/docs/_posts/Cabir40/2023-01-30-t5_lewip_informal_en.md new file mode 100644 index 00000000000000..49890ce449d62d --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_lewip_informal_en.md @@ -0,0 +1,83 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from SkolkovoInstitute) +author: John Snow Labs +name: t5_lewip_informal +date: 2023-01-30 +tags: [en, open_source, t5] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `LEWIP-informal` is a English model originally trained by `SkolkovoInstitute`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_lewip_informal_en_4.3.0_3.0_1675098112375.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_lewip_informal_en_4.3.0_3.0_1675098112375.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_lewip_informal","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_lewip_informal","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_lewip_informal| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|914.1 MB| + +## References + +- https://huggingface.co/SkolkovoInstitute/LEWIP-informal \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_logisgenerator_en.md b/docs/_posts/Cabir40/2023-01-30-t5_logisgenerator_en.md new file mode 100644 index 00000000000000..e7709c8b7881a9 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_logisgenerator_en.md @@ -0,0 +1,93 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from OnsElleuch) +author: John Snow Labs +name: t5_logisgenerator +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `logisgenerator` is a English model originally trained by `OnsElleuch`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_logisgenerator_en_4.3.0_3.0_1675104908400.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_logisgenerator_en_4.3.0_3.0_1675104908400.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_logisgenerator","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_logisgenerator","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_logisgenerator| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|280.2 MB| + +## References + +- https://huggingface.co/OnsElleuch/logisgenerator +- https://pypi.org/project/keytotext/ +- https://pepy.tech/project/keytotext +- https://colab.research.google.com/github/gagan3012/keytotext/blob/master/notebooks/K2T.ipynb +- https://share.streamlit.io/gagan3012/keytotext/UI/app.py +- https://github.com/gagan3012/keytotext#api +- https://hub.docker.com/r/gagan30/keytotext +- https://keytotext.readthedocs.io/en/latest/?badge=latest +- https://github.com/psf/black +- https://socialify.git.ci/gagan3012/keytotext/image?description=1&forks=1&language=1&owner=1&stargazers=1&theme=Light \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_longlm_base_zh.md b/docs/_posts/Cabir40/2023-01-30-t5_longlm_base_zh.md new file mode 100644 index 00000000000000..c259bdc0468c94 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_longlm_base_zh.md @@ -0,0 +1,85 @@ +--- +layout: model +title: Chinese T5ForConditionalGeneration Base Cased model (from thu-coai) +author: John Snow Labs +name: t5_longlm_base +date: 2023-01-30 +tags: [zh, open_source, t5] +task: Text Generation +language: zh +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `LongLM-base` is a Chinese model originally trained by `thu-coai`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_longlm_base_zh_4.3.0_3.0_1675098192540.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_longlm_base_zh_4.3.0_3.0_1675098192540.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_longlm_base","zh") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_longlm_base","zh") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_longlm_base| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|zh| +|Size:|926.8 MB| + +## References + +- https://huggingface.co/thu-coai/LongLM-base +- https://jianguanthu.github.io/ +- http://coai.cs.tsinghua.edu.cn/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_longlm_small_zh.md b/docs/_posts/Cabir40/2023-01-30-t5_longlm_small_zh.md new file mode 100644 index 00000000000000..a2e96d9f69bbb8 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_longlm_small_zh.md @@ -0,0 +1,86 @@ +--- +layout: model +title: Chinese T5ForConditionalGeneration Small Cased model (from thu-coai) +author: John Snow Labs +name: t5_longlm_small +date: 2023-01-30 +tags: [zh, open_source, t5, tensorflow] +task: Text Generation +language: zh +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `LongLM-small` is a Chinese model originally trained by `thu-coai`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_longlm_small_zh_4.3.0_3.0_1675098245048.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_longlm_small_zh_4.3.0_3.0_1675098245048.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_longlm_small","zh") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_longlm_small","zh") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_longlm_small| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|zh| +|Size:|288.2 MB| + +## References + +- https://huggingface.co/thu-coai/LongLM-small +- https://jianguanthu.github.io/ +- http://coai.cs.tsinghua.edu.cn/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_mengzi_base_chinese_correction_zh.md b/docs/_posts/Cabir40/2023-01-30-t5_mengzi_base_chinese_correction_zh.md new file mode 100644 index 00000000000000..8c788f5c9515cb --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_mengzi_base_chinese_correction_zh.md @@ -0,0 +1,89 @@ +--- +layout: model +title: Chinese T5ForConditionalGeneration Base Cased model (from shibing624) +author: John Snow Labs +name: t5_mengzi_base_chinese_correction +date: 2023-01-30 +tags: [zh, open_source, t5, tensorflow] +task: Text Generation +language: zh +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `mengzi-t5-base-chinese-correction` is a Chinese model originally trained by `shibing624`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_mengzi_base_chinese_correction_zh_4.3.0_3.0_1675105223361.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_mengzi_base_chinese_correction_zh_4.3.0_3.0_1675105223361.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_mengzi_base_chinese_correction","zh") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_mengzi_base_chinese_correction","zh") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_mengzi_base_chinese_correction| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|zh| +|Size:|1.0 GB| + +## References + +- https://huggingface.co/shibing624/mengzi-t5-base-chinese-correction +- https://github.com/shibing624/pycorrector +- https://github.com/shibing624/pycorrector/tree/master/pycorrector/t5 +- https://pan.baidu.com/s/1BV5tr9eONZCI0wERFvr0gQ +- http://nlp.ee.ncu.edu.tw/resource/csc.html +- https://github.com/wdimmy/Automatic-Corpus-Generation/blob/master/corpus/train.sgml \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_mengzi_base_zh.md b/docs/_posts/Cabir40/2023-01-30-t5_mengzi_base_zh.md new file mode 100644 index 00000000000000..5b86bc75703cbe --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_mengzi_base_zh.md @@ -0,0 +1,85 @@ +--- +layout: model +title: Chinese T5ForConditionalGeneration Base Cased model (from Langboat) +author: John Snow Labs +name: t5_mengzi_base +date: 2023-01-30 +tags: [zh, open_source, t5, tensorflow] +task: Text Generation +language: zh +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `mengzi-t5-base` is a Chinese model originally trained by `Langboat`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_mengzi_base_zh_4.3.0_3.0_1675105125486.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_mengzi_base_zh_4.3.0_3.0_1675105125486.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_mengzi_base","zh") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_mengzi_base","zh") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_mengzi_base| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|zh| +|Size:|1.0 GB| + +## References + +- https://huggingface.co/Langboat/mengzi-t5-base +- https://arxiv.org/abs/2110.06696 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_mengzit5_comment_zh.md b/docs/_posts/Cabir40/2023-01-30-t5_mengzit5_comment_zh.md new file mode 100644 index 00000000000000..b1eb7fe30dd2b1 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_mengzit5_comment_zh.md @@ -0,0 +1,84 @@ +--- +layout: model +title: Chinese T5ForConditionalGeneration Cased model (from wawaup) +author: John Snow Labs +name: t5_mengzit5_comment +date: 2023-01-30 +tags: [zh, open_source, t5] +task: Text Generation +language: zh +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `MengziT5-Comment` is a Chinese model originally trained by `wawaup`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_mengzit5_comment_zh_4.3.0_3.0_1675098308627.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_mengzit5_comment_zh_4.3.0_3.0_1675098308627.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_mengzit5_comment","zh") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_mengzit5_comment","zh") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_mengzit5_comment| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|zh| +|Size:|1.0 GB| + +## References + +- https://huggingface.co/wawaup/MengziT5-Comment +- https://github.com/lancopku/Graph-to-seq-comment-generation \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_mixqg_base_en.md b/docs/_posts/Cabir40/2023-01-30-t5_mixqg_base_en.md new file mode 100644 index 00000000000000..a5786368165b30 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_mixqg_base_en.md @@ -0,0 +1,86 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from Salesforce) +author: John Snow Labs +name: t5_mixqg_base +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `mixqg-base` is a English model originally trained by `Salesforce`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_mixqg_base_en_4.3.0_3.0_1675105642190.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_mixqg_base_en_4.3.0_3.0_1675105642190.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_mixqg_base","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_mixqg_base","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_mixqg_base| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|583.1 MB| + +## References + +- https://huggingface.co/Salesforce/mixqg-base +- https://arxiv.org/abs/2110.08175 +- https://github.com/salesforce/QGen \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_msmarco_base_v1_en.md b/docs/_posts/Cabir40/2023-01-30-t5_msmarco_base_v1_en.md new file mode 100644 index 00000000000000..756f9fe59e9283 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_msmarco_base_v1_en.md @@ -0,0 +1,90 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from doc2query) +author: John Snow Labs +name: t5_msmarco_base_v1 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `msmarco-t5-base-v1` is a English model originally trained by `doc2query`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_msmarco_base_v1_en_4.3.0_3.0_1675105723928.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_msmarco_base_v1_en_4.3.0_3.0_1675105723928.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_msmarco_base_v1","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_msmarco_base_v1","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_msmarco_base_v1| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|1.0 GB| + +## References + +- https://huggingface.co/doc2query/msmarco-t5-base-v1 +- https://arxiv.org/abs/1904.08375 +- https://cs.uwaterloo.ca/~jimmylin/publications/Nogueira_Lin_2019_docTTTTTquery-v2.pdf +- https://arxiv.org/abs/2104.08663 +- https://github.com/UKPLab/beir +- https://www.sbert.net/examples/unsupervised_learning/query_generation/README.html +- https://github.com/microsoft/MSMARCO-Passage-Ranking \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_msmarco_small_v1_en.md b/docs/_posts/Cabir40/2023-01-30-t5_msmarco_small_v1_en.md new file mode 100644 index 00000000000000..9d31d65fdea362 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_msmarco_small_v1_en.md @@ -0,0 +1,90 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from doc2query) +author: John Snow Labs +name: t5_msmarco_small_v1 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `msmarco-t5-small-v1` is a English model originally trained by `doc2query`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_msmarco_small_v1_en_4.3.0_3.0_1675105793797.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_msmarco_small_v1_en_4.3.0_3.0_1675105793797.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_msmarco_small_v1","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_msmarco_small_v1","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_msmarco_small_v1| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|340.7 MB| + +## References + +- https://huggingface.co/doc2query/msmarco-t5-small-v1 +- https://arxiv.org/abs/1904.08375 +- https://cs.uwaterloo.ca/~jimmylin/publications/Nogueira_Lin_2019_docTTTTTquery-v2.pdf +- https://arxiv.org/abs/2104.08663 +- https://github.com/UKPLab/beir +- https://www.sbert.net/examples/unsupervised_learning/query_generation/README.html +- https://github.com/microsoft/MSMARCO-Passage-Ranking \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_mt5_base_it_paraphraser_it.md b/docs/_posts/Cabir40/2023-01-30-t5_mt5_base_it_paraphraser_it.md new file mode 100644 index 00000000000000..8b437c1de913a6 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_mt5_base_it_paraphraser_it.md @@ -0,0 +1,89 @@ +--- +layout: model +title: Italian T5ForConditionalGeneration Base Cased model (from aiknowyou) +author: John Snow Labs +name: t5_mt5_base_it_paraphraser +date: 2023-01-30 +tags: [it, open_source, t5, tensorflow] +task: Text Generation +language: it +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `mt5-base-it-paraphraser` is a Italian model originally trained by `aiknowyou`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_mt5_base_it_paraphraser_it_4.3.0_3.0_1675105866508.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_mt5_base_it_paraphraser_it_4.3.0_3.0_1675105866508.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_mt5_base_it_paraphraser","it") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_mt5_base_it_paraphraser","it") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_mt5_base_it_paraphraser| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|it| +|Size:|969.5 MB| + +## References + +- https://huggingface.co/aiknowyou/mt5-base-it-paraphraser +- https://arxiv.org/abs/2010.11934 +- https://colab.research.google.com/drive/1DGeF190gJ3DjRFQiwhFuZalp427iqJNQ +- https://gist.github.com/avidale/44cd35bfcdaf8bedf51d97c468cc8001 +- http://creativecommons.org/licenses/by-nc-sa/4.0/ +- http://creativecommons.org/licenses/by-nc-sa/4.0/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_mt5_base_sinaha_qa_si.md b/docs/_posts/Cabir40/2023-01-30-t5_mt5_base_sinaha_qa_si.md new file mode 100644 index 00000000000000..24780be860d21e --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_mt5_base_sinaha_qa_si.md @@ -0,0 +1,84 @@ +--- +layout: model +title: Sinhala T5ForConditionalGeneration Base Cased model (from sankhajay) +author: John Snow Labs +name: t5_mt5_base_sinaha_qa +date: 2023-01-30 +tags: [si, open_source, t5, tensorflow] +task: Text Generation +language: si +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `mt5-base-sinaha-qa` is a Sinhala model originally trained by `sankhajay`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_mt5_base_sinaha_qa_si_4.3.0_3.0_1675106104485.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_mt5_base_sinaha_qa_si_4.3.0_3.0_1675106104485.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_mt5_base_sinaha_qa","si") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_mt5_base_sinaha_qa","si") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_mt5_base_sinaha_qa| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|si| +|Size:|1.2 GB| + +## References + +- https://huggingface.co/sankhajay/mt5-base-sinaha-qa \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_mt5_small_german_finetune_mlsum_de.md b/docs/_posts/Cabir40/2023-01-30-t5_mt5_small_german_finetune_mlsum_de.md new file mode 100644 index 00000000000000..2a30b3d9cfb618 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_mt5_small_german_finetune_mlsum_de.md @@ -0,0 +1,85 @@ +--- +layout: model +title: German T5ForConditionalGeneration Small Cased model (from ml6team) +author: John Snow Labs +name: t5_mt5_small_german_finetune_mlsum +date: 2023-01-30 +tags: [de, open_source, t5, tensorflow] +task: Text Generation +language: de +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `mt5-small-german-finetune-mlsum` is a German model originally trained by `ml6team`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_mt5_small_german_finetune_mlsum_de_4.3.0_3.0_1675106313224.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_mt5_small_german_finetune_mlsum_de_4.3.0_3.0_1675106313224.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_mt5_small_german_finetune_mlsum","de") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_mt5_small_german_finetune_mlsum","de") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_mt5_small_german_finetune_mlsum| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|de| +|Size:|1.3 GB| + +## References + +- https://huggingface.co/ml6team/mt5-small-german-finetune-mlsum +- https://github.com/pltrdy/rouge \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_ner_conll_entityreplace_en.md b/docs/_posts/Cabir40/2023-01-30-t5_ner_conll_entityreplace_en.md new file mode 100644 index 00000000000000..ba3ad6d3715b7b --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_ner_conll_entityreplace_en.md @@ -0,0 +1,86 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from pitehu) +author: John Snow Labs +name: t5_ner_conll_entityreplace +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `T5_NER_CONLL_ENTITYREPLACE` is a English model originally trained by `pitehu`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_ner_conll_entityreplace_en_4.3.0_3.0_1675099568513.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_ner_conll_entityreplace_en_4.3.0_3.0_1675099568513.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_ner_conll_entityreplace","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_ner_conll_entityreplace","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_ner_conll_entityreplace| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|275.5 MB| + +## References + +- https://huggingface.co/pitehu/T5_NER_CONLL_ENTITYREPLACE +- https://arxiv.org/pdf/2111.10952.pdf +- https://arxiv.org/pdf/1810.04805.pdf \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_ner_conll_list_en.md b/docs/_posts/Cabir40/2023-01-30-t5_ner_conll_list_en.md new file mode 100644 index 00000000000000..fdcf2fdf23c778 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_ner_conll_list_en.md @@ -0,0 +1,84 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from pitehu) +author: John Snow Labs +name: t5_ner_conll_list +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `T5_NER_CONLL_LIST` is a English model originally trained by `pitehu`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_ner_conll_list_en_4.3.0_3.0_1675099601757.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_ner_conll_list_en_4.3.0_3.0_1675099601757.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_ner_conll_list","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_ner_conll_list","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_ner_conll_list| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|272.6 MB| + +## References + +- https://huggingface.co/pitehu/T5_NER_CONLL_LIST \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_neutrally_en.md b/docs/_posts/Cabir40/2023-01-30-t5_neutrally_en.md new file mode 100644 index 00000000000000..6c825187704ea9 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_neutrally_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from erickfm) +author: John Snow Labs +name: t5_neutrally +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `neutrally` is a English model originally trained by `erickfm`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_neutrally_en_4.3.0_3.0_1675106407099.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_neutrally_en_4.3.0_3.0_1675106407099.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_neutrally","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_neutrally","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_neutrally| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|923.9 MB| + +## References + +- https://huggingface.co/erickfm/neutrally +- https://github.com/rpryzant/neutralizing-bias +- https://nlp.stanford.edu/pubs/pryzant2020bias.pdf +- https://en.wikipedia.org/wiki/BLEU +- https://apps-summer22.ischool.berkeley.edu/neutrally/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_panggi_base_indonesian_summarization_cased_id.md b/docs/_posts/Cabir40/2023-01-30-t5_panggi_base_indonesian_summarization_cased_id.md new file mode 100644 index 00000000000000..881f706b676c9b --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_panggi_base_indonesian_summarization_cased_id.md @@ -0,0 +1,85 @@ +--- +layout: model +title: Indonesian T5ForConditionalGeneration Base Cased model (from panggi) +author: John Snow Labs +name: t5_panggi_base_indonesian_summarization_cased +date: 2023-01-30 +tags: [id, open_source, t5, tensorflow] +task: Text Generation +language: id +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-base-indonesian-summarization-cased` is a Indonesian model originally trained by `panggi`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_panggi_base_indonesian_summarization_cased_id_4.3.0_3.0_1675109758797.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_panggi_base_indonesian_summarization_cased_id_4.3.0_3.0_1675109758797.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_panggi_base_indonesian_summarization_cased","id") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_panggi_base_indonesian_summarization_cased","id") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_panggi_base_indonesian_summarization_cased| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|id| +|Size:|925.6 MB| + +## References + +- https://huggingface.co/panggi/t5-base-indonesian-summarization-cased +- https://github.com/kata-ai/indosum \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_plt5_base_pl.md b/docs/_posts/Cabir40/2023-01-30-t5_plt5_base_pl.md new file mode 100644 index 00000000000000..cd760cb76c30e0 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_plt5_base_pl.md @@ -0,0 +1,92 @@ +--- +layout: model +title: Polish T5ForConditionalGeneration Base Cased model (from allegro) +author: John Snow Labs +name: t5_plt5_base +date: 2023-01-30 +tags: [pl, open_source, t5, tensorflow] +task: Text Generation +language: pl +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `plt5-base` is a Polish model originally trained by `allegro`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_plt5_base_pl_4.3.0_3.0_1675106653699.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_plt5_base_pl_4.3.0_3.0_1675106653699.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_plt5_base","pl") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_plt5_base","pl") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_plt5_base| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|pl| +|Size:|601.0 MB| + +## References + +- https://huggingface.co/allegro/plt5-base +- https://github.com/facebookresearch/cc_net +- https://github.com/facebookresearch/cc_net +- http://nkjp.pl/index.php?page=14&lang=1 +- http://opus.nlpl.eu/OpenSubtitles-v2018.php +- https://dumps.wikimedia.org/ +- https://wolnelektury.pl/ +- https://ml.allegro.tech/ +- http://zil.ipipan.waw.pl/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_plt5_base_poquad_pl.md b/docs/_posts/Cabir40/2023-01-30-t5_plt5_base_poquad_pl.md new file mode 100644 index 00000000000000..0bf473e9877e83 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_plt5_base_poquad_pl.md @@ -0,0 +1,84 @@ +--- +layout: model +title: Polish T5ForConditionalGeneration Base Cased model (from azwierzc) +author: John Snow Labs +name: t5_plt5_base_poquad +date: 2023-01-30 +tags: [pl, open_source, t5, tensorflow] +task: Text Generation +language: pl +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `plt5-base-poquad` is a Polish model originally trained by `azwierzc`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_plt5_base_poquad_pl_4.3.0_3.0_1675106743524.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_plt5_base_poquad_pl_4.3.0_3.0_1675106743524.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_plt5_base_poquad","pl") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_plt5_base_poquad","pl") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_plt5_base_poquad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|pl| +|Size:|1.1 GB| + +## References + +- https://huggingface.co/azwierzc/plt5-base-poquad \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_randeng_77m_multitask_chinese_zh.md b/docs/_posts/Cabir40/2023-01-30-t5_randeng_77m_multitask_chinese_zh.md new file mode 100644 index 00000000000000..b0f79fcdfbf62b --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_randeng_77m_multitask_chinese_zh.md @@ -0,0 +1,97 @@ +--- +layout: model +title: Chinese T5ForConditionalGeneration Cased model (from IDEA-CCNL) +author: John Snow Labs +name: t5_randeng_77m_multitask_chinese +date: 2023-01-30 +tags: [zh, open_source, t5, tensorflow] +task: Text Generation +language: zh +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `Randeng-T5-77M-MultiTask-Chinese` is a Chinese model originally trained by `IDEA-CCNL`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_randeng_77m_multitask_chinese_zh_4.3.0_3.0_1675098367899.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_randeng_77m_multitask_chinese_zh_4.3.0_3.0_1675098367899.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_randeng_77m_multitask_chinese","zh") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_randeng_77m_multitask_chinese","zh") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_randeng_77m_multitask_chinese| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|zh| +|Size:|349.2 MB| + +## References + +- https://huggingface.co/IDEA-CCNL/Randeng-T5-77M-MultiTask-Chinese +- https://github.com/IDEA-CCNL/Fengshenbang-LM +- https://fengshenbang-doc.readthedocs.io/ +- http://jmlr.org/papers/v21/20-074.html +- https://github.com/IDEA-CCNL/Fengshenbang-LM/ +- https://github.com/IDEA-CCNL/Fengshenbang-LM/tree/main/fengshen/examples/pretrain_t5 +- https://github.com/IDEA-CCNL/Fengshenbang-LM/tree/main/fengshen/examples/mt5_summary +- https://github.com/IDEA-CCNL/Fengshenbang-LM/ +- https://github.com/IDEA-CCNL/Fengshenbang-LM/tree/main/fengshen/examples/pretrain_t5 +- https://github.com/IDEA-CCNL/Fengshenbang-LM/tree/main/fengshen/examples/mt5_summary +- https://arxiv.org/abs/2209.02970 +- https://arxiv.org/abs/2209.02970 +- https://github.com/IDEA-CCNL/Fengshenbang-LM/ +- https://github.com/IDEA-CCNL/Fengshenbang-LM/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_rut5_base_headline_gen_telegram_ru.md b/docs/_posts/Cabir40/2023-01-30-t5_rut5_base_headline_gen_telegram_ru.md new file mode 100644 index 00000000000000..3a5b85c1bb7b35 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_rut5_base_headline_gen_telegram_ru.md @@ -0,0 +1,86 @@ +--- +layout: model +title: Russian T5ForConditionalGeneration Base Cased model (from IlyaGusev) +author: John Snow Labs +name: t5_rut5_base_headline_gen_telegram +date: 2023-01-30 +tags: [ru, open_source, t5, tensorflow] +task: Text Generation +language: ru +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `rut5_base_headline_gen_telegram` is a Russian model originally trained by `IlyaGusev`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_rut5_base_headline_gen_telegram_ru_4.3.0_3.0_1675106899958.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_rut5_base_headline_gen_telegram_ru_4.3.0_3.0_1675106899958.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_rut5_base_headline_gen_telegram","ru") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_rut5_base_headline_gen_telegram","ru") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_rut5_base_headline_gen_telegram| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|ru| +|Size:|995.4 MB| + +## References + +- https://huggingface.co/IlyaGusev/rut5_base_headline_gen_telegram +- https://www.dropbox.com/s/ykqk49a8avlmnaf/ru_all_split.tar.gz +- https://github.com/IlyaGusev/summarus/blob/master/external/hf_scripts/train.py \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_rut5_base_sum_gazeta_ru.md b/docs/_posts/Cabir40/2023-01-30-t5_rut5_base_sum_gazeta_ru.md new file mode 100644 index 00000000000000..83589297b8b4c4 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_rut5_base_sum_gazeta_ru.md @@ -0,0 +1,88 @@ +--- +layout: model +title: Russian T5ForConditionalGeneration Base Cased model (from IlyaGusev) +author: John Snow Labs +name: t5_rut5_base_sum_gazeta +date: 2023-01-30 +tags: [ru, open_source, t5, tensorflow] +task: Text Generation +language: ru +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `rut5_base_sum_gazeta` is a Russian model originally trained by `IlyaGusev`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_rut5_base_sum_gazeta_ru_4.3.0_3.0_1675106989342.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_rut5_base_sum_gazeta_ru_4.3.0_3.0_1675106989342.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_rut5_base_sum_gazeta","ru") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_rut5_base_sum_gazeta","ru") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_rut5_base_sum_gazeta| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|ru| +|Size:|991.7 MB| + +## References + +- https://huggingface.co/IlyaGusev/rut5_base_sum_gazeta +- https://colab.research.google.com/drive/1re5E26ZIDUpAx1gOCZkbF3hcwjozmgG0 +- https://github.com/IlyaGusev/summarus/blob/master/external/hf_scripts/train.py +- https://github.com/IlyaGusev/summarus/blob/master/external/hf_scripts/configs/t5_training_config.json +- https://github.com/IlyaGusev/summarus/blob/master/evaluate.py \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_rut5_small_chitchat_ru.md b/docs/_posts/Cabir40/2023-01-30-t5_rut5_small_chitchat_ru.md new file mode 100644 index 00000000000000..cc0ef61a27be21 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_rut5_small_chitchat_ru.md @@ -0,0 +1,84 @@ +--- +layout: model +title: Russian T5ForConditionalGeneration Small Cased model (from cointegrated) +author: John Snow Labs +name: t5_rut5_small_chitchat +date: 2023-01-30 +tags: [ru, open_source, t5, tensorflow] +task: Text Generation +language: ru +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `rut5-small-chitchat` is a Russian model originally trained by `cointegrated`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_rut5_small_chitchat_ru_4.3.0_3.0_1675106805725.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_rut5_small_chitchat_ru_4.3.0_3.0_1675106805725.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_rut5_small_chitchat","ru") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_rut5_small_chitchat","ru") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_rut5_small_chitchat| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|ru| +|Size:|277.4 MB| + +## References + +- https://huggingface.co/cointegrated/rut5-small-chitchat \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_rut5_small_normalizer_ru.md b/docs/_posts/Cabir40/2023-01-30-t5_rut5_small_normalizer_ru.md new file mode 100644 index 00000000000000..221756ef4aafc2 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_rut5_small_normalizer_ru.md @@ -0,0 +1,87 @@ +--- +layout: model +title: Russian T5ForConditionalGeneration Small Cased model (from cointegrated) +author: John Snow Labs +name: t5_rut5_small_normalizer +date: 2023-01-30 +tags: [ru, open_source, t5, tensorflow] +task: Text Generation +language: ru +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `rut5-small-normalizer` is a Russian model originally trained by `cointegrated`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_rut5_small_normalizer_ru_4.3.0_3.0_1675106835622.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_rut5_small_normalizer_ru_4.3.0_3.0_1675106835622.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_rut5_small_normalizer","ru") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_rut5_small_normalizer","ru") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_rut5_small_normalizer| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|ru| +|Size:|277.8 MB| + +## References + +- https://huggingface.co/cointegrated/rut5-small-normalizer +- https://github.com/natasha/natasha +- https://github.com/kmike/pymorphy2 +- https://wortschatz.uni-leipzig.de/en/download/Russian \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_rut5_tox_ru.md b/docs/_posts/Cabir40/2023-01-30-t5_rut5_tox_ru.md new file mode 100644 index 00000000000000..587c9bf131571b --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_rut5_tox_ru.md @@ -0,0 +1,84 @@ +--- +layout: model +title: Russian T5ForConditionalGeneration Cased model (from IlyaGusev) +author: John Snow Labs +name: t5_rut5_tox +date: 2023-01-30 +tags: [ru, open_source, t5, tensorflow] +task: Text Generation +language: ru +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `rut5_tox` is a Russian model originally trained by `IlyaGusev`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_rut5_tox_ru_4.3.0_3.0_1675107092843.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_rut5_tox_ru_4.3.0_3.0_1675107092843.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_rut5_tox","ru") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_rut5_tox","ru") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_rut5_tox| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|ru| +|Size:|955.7 MB| + +## References + +- https://huggingface.co/IlyaGusev/rut5_tox \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_s2orc_base_v1_en.md b/docs/_posts/Cabir40/2023-01-30-t5_s2orc_base_v1_en.md new file mode 100644 index 00000000000000..e11fe51446820f --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_s2orc_base_v1_en.md @@ -0,0 +1,89 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from doc2query) +author: John Snow Labs +name: t5_s2orc_base_v1 +date: 2023-01-30 +tags: [en, open_source, t5] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `S2ORC-t5-base-v1` is a English model originally trained by `doc2query`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_s2orc_base_v1_en_4.3.0_3.0_1675098438086.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_s2orc_base_v1_en_4.3.0_3.0_1675098438086.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_s2orc_base_v1","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_s2orc_base_v1","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_s2orc_base_v1| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|1.0 GB| + +## References + +- https://huggingface.co/doc2query/S2ORC-t5-base-v1 +- https://arxiv.org/abs/1904.08375 +- https://cs.uwaterloo.ca/~jimmylin/publications/Nogueira_Lin_2019_docTTTTTquery-v2.pdf +- https://arxiv.org/abs/2104.08663 +- https://github.com/UKPLab/beir +- https://www.sbert.net/examples/unsupervised_learning/query_generation/README.html +- https://github.com/allenai/s2orc \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_sber_rut5_filler_ru.md b/docs/_posts/Cabir40/2023-01-30-t5_sber_rut5_filler_ru.md new file mode 100644 index 00000000000000..790d6c2fc75dfa --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_sber_rut5_filler_ru.md @@ -0,0 +1,84 @@ +--- +layout: model +title: Russian T5ForConditionalGeneration Cased model (from IlyaGusev) +author: John Snow Labs +name: t5_sber_rut5_filler +date: 2023-01-30 +tags: [ru, open_source, t5, tensorflow] +task: Text Generation +language: ru +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `sber_rut5_filler` is a Russian model originally trained by `IlyaGusev`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_sber_rut5_filler_ru_4.3.0_3.0_1675107176054.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_sber_rut5_filler_ru_4.3.0_3.0_1675107176054.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_sber_rut5_filler","ru") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_sber_rut5_filler","ru") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_sber_rut5_filler| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|ru| +|Size:|927.4 MB| + +## References + +- https://huggingface.co/IlyaGusev/sber_rut5_filler \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_scifive_base_pmc_en.md b/docs/_posts/Cabir40/2023-01-30-t5_scifive_base_pmc_en.md new file mode 100644 index 00000000000000..09b0868d30c92f --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_scifive_base_pmc_en.md @@ -0,0 +1,85 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from razent) +author: John Snow Labs +name: t5_scifive_base_pmc +date: 2023-01-30 +tags: [en, open_source, t5] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `SciFive-base-PMC` is a English model originally trained by `razent`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_scifive_base_pmc_en_4.3.0_3.0_1675098735326.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_scifive_base_pmc_en_4.3.0_3.0_1675098735326.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_scifive_base_pmc","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_scifive_base_pmc","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_scifive_base_pmc| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|474.3 MB| + +## References + +- https://huggingface.co/razent/SciFive-base-PMC +- https://arxiv.org/abs/2106.03598 +- https://github.com/justinphan3110/SciFive \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_scifive_base_pubmed_en.md b/docs/_posts/Cabir40/2023-01-30-t5_scifive_base_pubmed_en.md new file mode 100644 index 00000000000000..5e0fd6389f9ee8 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_scifive_base_pubmed_en.md @@ -0,0 +1,86 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from razent) +author: John Snow Labs +name: t5_scifive_base_pubmed +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `SciFive-base-Pubmed` is a English model originally trained by `razent`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_scifive_base_pubmed_en_4.3.0_3.0_1675098922792.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_scifive_base_pubmed_en_4.3.0_3.0_1675098922792.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_scifive_base_pubmed","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_scifive_base_pubmed","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_scifive_base_pubmed| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|474.3 MB| + +## References + +- https://huggingface.co/razent/SciFive-base-Pubmed +- https://arxiv.org/abs/2106.03598 +- https://github.com/justinphan3110/SciFive \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_scifive_base_pubmed_pmc_en.md b/docs/_posts/Cabir40/2023-01-30-t5_scifive_base_pubmed_pmc_en.md new file mode 100644 index 00000000000000..a67f15dd21490c --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_scifive_base_pubmed_pmc_en.md @@ -0,0 +1,86 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from razent) +author: John Snow Labs +name: t5_scifive_base_pubmed_pmc +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `SciFive-base-Pubmed_PMC` is a English model originally trained by `razent`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_scifive_base_pubmed_pmc_en_4.3.0_3.0_1675099109583.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_scifive_base_pubmed_pmc_en_4.3.0_3.0_1675099109583.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_scifive_base_pubmed_pmc","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_scifive_base_pubmed_pmc","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_scifive_base_pubmed_pmc| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|474.3 MB| + +## References + +- https://huggingface.co/razent/SciFive-base-Pubmed_PMC +- https://arxiv.org/abs/2106.03598 +- https://github.com/justinphan3110/SciFive \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_ssr_base_en.md b/docs/_posts/Cabir40/2023-01-30-t5_ssr_base_en.md new file mode 100644 index 00000000000000..0fc9d549215fec --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_ssr_base_en.md @@ -0,0 +1,85 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from microsoft) +author: John Snow Labs +name: t5_ssr_base +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `ssr-base` is a English model originally trained by `microsoft`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_ssr_base_en_4.3.0_3.0_1675107262685.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_ssr_base_en_4.3.0_3.0_1675107262685.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_ssr_base","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_ssr_base","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_ssr_base| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|926.9 MB| + +## References + +- https://huggingface.co/microsoft/ssr-base +- https://arxiv.org/abs/2101.00416 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_stackexchange_base_v1_en.md b/docs/_posts/Cabir40/2023-01-30-t5_stackexchange_base_v1_en.md new file mode 100644 index 00000000000000..46b137aeb82236 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_stackexchange_base_v1_en.md @@ -0,0 +1,89 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from doc2query) +author: John Snow Labs +name: t5_stackexchange_base_v1 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `stackexchange-t5-base-v1` is a English model originally trained by `doc2query`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_stackexchange_base_v1_en_4.3.0_3.0_1675107351420.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_stackexchange_base_v1_en_4.3.0_3.0_1675107351420.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_stackexchange_base_v1","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_stackexchange_base_v1","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_stackexchange_base_v1| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|1.0 GB| + +## References + +- https://huggingface.co/doc2query/stackexchange-t5-base-v1 +- https://arxiv.org/abs/1904.08375 +- https://cs.uwaterloo.ca/~jimmylin/publications/Nogueira_Lin_2019_docTTTTTquery-v2.pdf +- https://arxiv.org/abs/2104.08663 +- https://github.com/UKPLab/beir +- https://www.sbert.net/examples/unsupervised_learning/query_generation/README.html \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_stackexchange_title_body_base_v1_en.md b/docs/_posts/Cabir40/2023-01-30-t5_stackexchange_title_body_base_v1_en.md new file mode 100644 index 00000000000000..5be0004f63dabe --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_stackexchange_title_body_base_v1_en.md @@ -0,0 +1,89 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from doc2query) +author: John Snow Labs +name: t5_stackexchange_title_body_base_v1 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `stackexchange-title-body-t5-base-v1` is a English model originally trained by `doc2query`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_stackexchange_title_body_base_v1_en_4.3.0_3.0_1675107445767.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_stackexchange_title_body_base_v1_en_4.3.0_3.0_1675107445767.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_stackexchange_title_body_base_v1","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_stackexchange_title_body_base_v1","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_stackexchange_title_body_base_v1| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|1.0 GB| + +## References + +- https://huggingface.co/doc2query/stackexchange-title-body-t5-base-v1 +- https://arxiv.org/abs/1904.08375 +- https://cs.uwaterloo.ca/~jimmylin/publications/Nogueira_Lin_2019_docTTTTTquery-v2.pdf +- https://arxiv.org/abs/2104.08663 +- https://github.com/UKPLab/beir +- https://www.sbert.net/examples/unsupervised_learning/query_generation/README.html \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_stackexchange_title_body_small_v1_en.md b/docs/_posts/Cabir40/2023-01-30-t5_stackexchange_title_body_small_v1_en.md new file mode 100644 index 00000000000000..7d93aeaeac4684 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_stackexchange_title_body_small_v1_en.md @@ -0,0 +1,89 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from doc2query) +author: John Snow Labs +name: t5_stackexchange_title_body_small_v1 +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `stackexchange-title-body-t5-small-v1` is a English model originally trained by `doc2query`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_stackexchange_title_body_small_v1_en_4.3.0_3.0_1675107510591.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_stackexchange_title_body_small_v1_en_4.3.0_3.0_1675107510591.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_stackexchange_title_body_small_v1","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_stackexchange_title_body_small_v1","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_stackexchange_title_body_small_v1| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|347.9 MB| + +## References + +- https://huggingface.co/doc2query/stackexchange-title-body-t5-small-v1 +- https://arxiv.org/abs/1904.08375 +- https://cs.uwaterloo.ca/~jimmylin/publications/Nogueira_Lin_2019_docTTTTTquery-v2.pdf +- https://arxiv.org/abs/2104.08663 +- https://github.com/UKPLab/beir +- https://www.sbert.net/examples/unsupervised_learning/query_generation/README.html \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_stekgramarchecker_en.md b/docs/_posts/Cabir40/2023-01-30-t5_stekgramarchecker_en.md new file mode 100644 index 00000000000000..c52fb33d160f2c --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_stekgramarchecker_en.md @@ -0,0 +1,86 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from Neo87z1) +author: John Snow Labs +name: t5_stekgramarchecker +date: 2023-01-30 +tags: [en, open_source, t5] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `STEKGramarChecker` is a English model originally trained by `Neo87z1`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_stekgramarchecker_en_4.3.0_3.0_1675098528002.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_stekgramarchecker_en_4.3.0_3.0_1675098528002.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_stekgramarchecker","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_stekgramarchecker","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_stekgramarchecker| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|904.6 MB| + +## References + +- https://huggingface.co/Neo87z1/STEKGramarChecker +- https://github.com/EricFillion/happy-transformer +- https://arxiv.org/abs/1702.04066 +- https://www.vennify.ai/fine-tune-grammar-correction/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_stocks_news_en.md b/docs/_posts/Cabir40/2023-01-30-t5_stocks_news_en.md new file mode 100644 index 00000000000000..df7d15cd15efea --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_stocks_news_en.md @@ -0,0 +1,85 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from cometrain) +author: John Snow Labs +name: t5_stocks_news +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `stocks-news-t5` is a English model originally trained by `cometrain`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_stocks_news_en_4.3.0_3.0_1675107545220.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_stocks_news_en_4.3.0_3.0_1675107545220.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_stocks_news","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_stocks_news","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_stocks_news| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|277.2 MB| + +## References + +- https://huggingface.co/cometrain/stocks-news-t5 +- https://www.kaggle.com/datasets/sbhatti/financial-sentiment-analysis \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_t2t_adex_prompt_en.md b/docs/_posts/Cabir40/2023-01-30-t5_t2t_adex_prompt_en.md new file mode 100644 index 00000000000000..3431742781eb59 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_t2t_adex_prompt_en.md @@ -0,0 +1,85 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from gokceuludogan) +author: John Snow Labs +name: t5_t2t_adex_prompt +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t2t-adeX-prompt` is a English model originally trained by `gokceuludogan`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_t2t_adex_prompt_en_4.3.0_3.0_1675107607187.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_t2t_adex_prompt_en_4.3.0_3.0_1675107607187.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_t2t_adex_prompt","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_t2t_adex_prompt","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_t2t_adex_prompt| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|925.3 MB| + +## References + +- https://huggingface.co/gokceuludogan/t2t-adeX-prompt +- https://github.com/gokceuludogan/boun-tabi-smm4h22 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_t2t_assert_ade_balanced_en.md b/docs/_posts/Cabir40/2023-01-30-t5_t2t_assert_ade_balanced_en.md new file mode 100644 index 00000000000000..4c5d1892896920 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_t2t_assert_ade_balanced_en.md @@ -0,0 +1,85 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from yirmibesogluz) +author: John Snow Labs +name: t5_t2t_assert_ade_balanced +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t2t-assert-ade-balanced` is a English model originally trained by `yirmibesogluz`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_t2t_assert_ade_balanced_en_4.3.0_3.0_1675107688482.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_t2t_assert_ade_balanced_en_4.3.0_3.0_1675107688482.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_t2t_assert_ade_balanced","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_t2t_assert_ade_balanced","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_t2t_assert_ade_balanced| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|917.2 MB| + +## References + +- https://huggingface.co/yirmibesogluz/t2t-assert-ade-balanced +- https://github.com/gokceuludogan/boun-tabi-smm4h22 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_t2t_ner_ade_balanced_en.md b/docs/_posts/Cabir40/2023-01-30-t5_t2t_ner_ade_balanced_en.md new file mode 100644 index 00000000000000..555c04293dc33a --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_t2t_ner_ade_balanced_en.md @@ -0,0 +1,85 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from yirmibesogluz) +author: John Snow Labs +name: t5_t2t_ner_ade_balanced +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t2t-ner-ade-balanced` is a English model originally trained by `yirmibesogluz`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_t2t_ner_ade_balanced_en_4.3.0_3.0_1675107775759.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_t2t_ner_ade_balanced_en_4.3.0_3.0_1675107775759.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_t2t_ner_ade_balanced","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_t2t_ner_ade_balanced","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_t2t_ner_ade_balanced| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|924.7 MB| + +## References + +- https://huggingface.co/yirmibesogluz/t2t-ner-ade-balanced +- https://github.com/gokceuludogan/boun-tabi-smm4h22 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_totto_base_bert_score_20k_steps_en.md b/docs/_posts/Cabir40/2023-01-30-t5_totto_base_bert_score_20k_steps_en.md new file mode 100644 index 00000000000000..dafc689e5eb829 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_totto_base_bert_score_20k_steps_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from Tejas21) +author: John Snow Labs +name: t5_totto_base_bert_score_20k_steps +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `Totto_t5_base_BERT_Score_20k_steps` is a English model originally trained by `Tejas21`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_totto_base_bert_score_20k_steps_en_4.3.0_3.0_1675099761482.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_totto_base_bert_score_20k_steps_en_4.3.0_3.0_1675099761482.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_totto_base_bert_score_20k_steps","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_totto_base_bert_score_20k_steps","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_totto_base_bert_score_20k_steps| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|925.2 MB| + +## References + +- https://huggingface.co/Tejas21/Totto_t5_base_BERT_Score_20k_steps +- https://github.com/google-research-datasets/ToTTo +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://github.com/google-research/language/tree/master/language/totto +- https://github.com/Tiiiger/bert_score \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-30-t5_ttparser_en.md b/docs/_posts/Cabir40/2023-01-30-t5_ttparser_en.md new file mode 100644 index 00000000000000..a7b6bc0ec4c68a --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-30-t5_ttparser_en.md @@ -0,0 +1,86 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from KES) +author: John Snow Labs +name: t5_ttparser +date: 2023-01-30 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `T5-TTParser` is a English model originally trained by `KES`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_ttparser_en_4.3.0_3.0_1675099426092.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_ttparser_en_4.3.0_3.0_1675099426092.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_ttparser","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_ttparser","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_ttparser| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|916.4 MB| + +## References + +- https://huggingface.co/KES/T5-TTParser +- https://arxiv.org/abs/1702.04066 +- https://pypi.org/project/Caribe/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_dl2_en.md b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_dl2_en.md new file mode 100644 index 00000000000000..7d2f6edc33431d --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_dl2_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Tiny Cased model (from google) +author: John Snow Labs +name: t5_efficient_tiny_dl2 +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-tiny-dl2` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_dl2_en_4.3.0_3.0_1675123232101.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_dl2_en_4.3.0_3.0_1675123232101.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_tiny_dl2","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_tiny_dl2","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_tiny_dl2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|54.2 MB| + +## References + +- https://huggingface.co/google/t5-efficient-tiny-dl2 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_dl6_en.md b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_dl6_en.md new file mode 100644 index 00000000000000..e153c123edd369 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_dl6_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Tiny Cased model (from google) +author: John Snow Labs +name: t5_efficient_tiny_dl6 +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-tiny-dl6` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_dl6_en_4.3.0_3.0_1675123262536.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_dl6_en_4.3.0_3.0_1675123262536.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_tiny_dl6","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_tiny_dl6","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_tiny_dl6| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|62.3 MB| + +## References + +- https://huggingface.co/google/t5-efficient-tiny-dl6 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_dl8_en.md b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_dl8_en.md new file mode 100644 index 00000000000000..a76def9988ceee --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_dl8_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Tiny Cased model (from google) +author: John Snow Labs +name: t5_efficient_tiny_dl8 +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-tiny-dl8` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_dl8_en_4.3.0_3.0_1675123295064.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_dl8_en_4.3.0_3.0_1675123295064.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_tiny_dl8","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_tiny_dl8","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_tiny_dl8| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|66.3 MB| + +## References + +- https://huggingface.co/google/t5-efficient-tiny-dl8 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_el12_en.md b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_el12_en.md new file mode 100644 index 00000000000000..9456addab4af42 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_el12_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Tiny Cased model (from google) +author: John Snow Labs +name: t5_efficient_tiny_el12 +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-tiny-el12` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_el12_en_4.3.0_3.0_1675123332949.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_el12_en_4.3.0_3.0_1675123332949.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_tiny_el12","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_tiny_el12","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_tiny_el12| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|74.4 MB| + +## References + +- https://huggingface.co/google/t5-efficient-tiny-el12 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_el2_en.md b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_el2_en.md new file mode 100644 index 00000000000000..e1cc1e2192c48c --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_el2_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Tiny Cased model (from google) +author: John Snow Labs +name: t5_efficient_tiny_el2 +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-tiny-el2` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_el2_en_4.3.0_3.0_1675123363126.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_el2_en_4.3.0_3.0_1675123363126.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_tiny_el2","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_tiny_el2","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_tiny_el2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|59.2 MB| + +## References + +- https://huggingface.co/google/t5-efficient-tiny-el2 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_el8_en.md b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_el8_en.md new file mode 100644 index 00000000000000..05ae09bee7a92a --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_el8_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Tiny Cased model (from google) +author: John Snow Labs +name: t5_efficient_tiny_el8 +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-tiny-el8` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_el8_en_4.3.0_3.0_1675123395876.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_el8_en_4.3.0_3.0_1675123395876.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_tiny_el8","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_tiny_el8","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_tiny_el8| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|68.3 MB| + +## References + +- https://huggingface.co/google/t5-efficient-tiny-el8 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_en.md b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_en.md new file mode 100644 index 00000000000000..a4b22420ca9c13 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Tiny Cased model (from google) +author: John Snow Labs +name: t5_efficient_tiny +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-tiny` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_en_4.3.0_3.0_1675123202788.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_en_4.3.0_3.0_1675123202788.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_tiny","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_tiny","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_tiny| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|46.1 MB| + +## References + +- https://huggingface.co/google/t5-efficient-tiny +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_ff12000_en.md b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_ff12000_en.md new file mode 100644 index 00000000000000..113c079fe6a1be --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_ff12000_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Tiny Cased model (from google) +author: John Snow Labs +name: t5_efficient_tiny_ff12000 +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-tiny-ff12000` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_ff12000_en_4.3.0_3.0_1675123452076.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_ff12000_en_4.3.0_3.0_1675123452076.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_tiny_ff12000","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_tiny_ff12000","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_tiny_ff12000| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|134.1 MB| + +## References + +- https://huggingface.co/google/t5-efficient-tiny-ff12000 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_ff2000_en.md b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_ff2000_en.md new file mode 100644 index 00000000000000..b72d9950fd67e9 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_ff2000_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Tiny Cased model (from google) +author: John Snow Labs +name: t5_efficient_tiny_ff2000 +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-tiny-ff2000` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_ff2000_en_4.3.0_3.0_1675123479854.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_ff2000_en_4.3.0_3.0_1675123479854.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_tiny_ff2000","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_tiny_ff2000","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_tiny_ff2000| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|46.1 MB| + +## References + +- https://huggingface.co/google/t5-efficient-tiny-ff2000 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_ff3000_en.md b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_ff3000_en.md new file mode 100644 index 00000000000000..54950e6f3e0d76 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_ff3000_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Tiny Cased model (from google) +author: John Snow Labs +name: t5_efficient_tiny_ff3000 +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-tiny-ff3000` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_ff3000_en_4.3.0_3.0_1675123510286.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_ff3000_en_4.3.0_3.0_1675123510286.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_tiny_ff3000","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_tiny_ff3000","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_tiny_ff3000| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|62.1 MB| + +## References + +- https://huggingface.co/google/t5-efficient-tiny-ff3000 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_ff6000_en.md b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_ff6000_en.md new file mode 100644 index 00000000000000..41469060b0877a --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_ff6000_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Tiny Cased model (from google) +author: John Snow Labs +name: t5_efficient_tiny_ff6000 +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-tiny-ff6000` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_ff6000_en_4.3.0_3.0_1675123549632.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_ff6000_en_4.3.0_3.0_1675123549632.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_tiny_ff6000","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_tiny_ff6000","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_tiny_ff6000| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|86.1 MB| + +## References + +- https://huggingface.co/google/t5-efficient-tiny-ff6000 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_ff9000_en.md b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_ff9000_en.md new file mode 100644 index 00000000000000..31289e533eb205 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_ff9000_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Tiny Cased model (from google) +author: John Snow Labs +name: t5_efficient_tiny_ff9000 +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-tiny-ff9000` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_ff9000_en_4.3.0_3.0_1675123598242.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_ff9000_en_4.3.0_3.0_1675123598242.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_tiny_ff9000","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_tiny_ff9000","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_tiny_ff9000| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|110.1 MB| + +## References + +- https://huggingface.co/google/t5-efficient-tiny-ff9000 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_nh16_en.md b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_nh16_en.md new file mode 100644 index 00000000000000..621332df7011b7 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_nh16_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Tiny Cased model (from google) +author: John Snow Labs +name: t5_efficient_tiny_nh16 +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-tiny-nh16` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_nh16_en_4.3.0_3.0_1675123654269.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_nh16_en_4.3.0_3.0_1675123654269.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_tiny_nh16","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_tiny_nh16","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_tiny_nh16| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|64.1 MB| + +## References + +- https://huggingface.co/google/t5-efficient-tiny-nh16 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_nh1_en.md b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_nh1_en.md new file mode 100644 index 00000000000000..c0d9f7d7a49f24 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_nh1_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Tiny Cased model (from google) +author: John Snow Labs +name: t5_efficient_tiny_nh1 +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-tiny-nh1` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_nh1_en_4.3.0_3.0_1675123623466.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_nh1_en_4.3.0_3.0_1675123623466.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_tiny_nh1","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_tiny_nh1","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_tiny_nh1| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|41.6 MB| + +## References + +- https://huggingface.co/google/t5-efficient-tiny-nh1 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_nh32_en.md b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_nh32_en.md new file mode 100644 index 00000000000000..13b978e1f9b553 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_nh32_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Tiny Cased model (from google) +author: John Snow Labs +name: t5_efficient_tiny_nh32 +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-tiny-nh32` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_nh32_en_4.3.0_3.0_1675123694521.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_nh32_en_4.3.0_3.0_1675123694521.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_tiny_nh32","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_tiny_nh32","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_tiny_nh32| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|88.2 MB| + +## References + +- https://huggingface.co/google/t5-efficient-tiny-nh32 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_nh8_en.md b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_nh8_en.md new file mode 100644 index 00000000000000..b7cbf356638de9 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_nh8_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Tiny Cased model (from google) +author: John Snow Labs +name: t5_efficient_tiny_nh8 +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-tiny-nh8` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_nh8_en_4.3.0_3.0_1675123720663.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_nh8_en_4.3.0_3.0_1675123720663.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_tiny_nh8","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_tiny_nh8","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_tiny_nh8| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|46.1 MB| + +## References + +- https://huggingface.co/google/t5-efficient-tiny-nh8 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_nl12_en.md b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_nl12_en.md new file mode 100644 index 00000000000000..3432ab35707f34 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_nl12_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Tiny Cased model (from google) +author: John Snow Labs +name: t5_efficient_tiny_nl12 +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-tiny-nl12` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_nl12_en_4.3.0_3.0_1675123754998.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_nl12_en_4.3.0_3.0_1675123754998.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_tiny_nl12","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_tiny_nl12","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_tiny_nl12| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|74.4 MB| + +## References + +- https://huggingface.co/google/t5-efficient-tiny-nl12 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_nl16_en.md b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_nl16_en.md new file mode 100644 index 00000000000000..9daeff2ae11ec8 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_nl16_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Tiny Cased model (from google) +author: John Snow Labs +name: t5_efficient_tiny_nl16 +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-tiny-nl16` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_nl16_en_4.3.0_3.0_1675123795068.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_nl16_en_4.3.0_3.0_1675123795068.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_tiny_nl16","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_tiny_nl16","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_tiny_nl16| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|88.6 MB| + +## References + +- https://huggingface.co/google/t5-efficient-tiny-nl16 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_nl24_en.md b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_nl24_en.md new file mode 100644 index 00000000000000..d393b49ec6bb16 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_nl24_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Tiny Cased model (from google) +author: John Snow Labs +name: t5_efficient_tiny_nl24 +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-tiny-nl24` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_nl24_en_4.3.0_3.0_1675123867646.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_nl24_en_4.3.0_3.0_1675123867646.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_tiny_nl24","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_tiny_nl24","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_tiny_nl24| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|117.0 MB| + +## References + +- https://huggingface.co/google/t5-efficient-tiny-nl24 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_nl2_en.md b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_nl2_en.md new file mode 100644 index 00000000000000..7fc8ffb6aeb61f --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_nl2_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Tiny Cased model (from google) +author: John Snow Labs +name: t5_efficient_tiny_nl2 +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-tiny-nl2` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_nl2_en_4.3.0_3.0_1675123819194.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_nl2_en_4.3.0_3.0_1675123819194.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_tiny_nl2","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_tiny_nl2","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_tiny_nl2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|39.0 MB| + +## References + +- https://huggingface.co/google/t5-efficient-tiny-nl2 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_nl32_en.md b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_nl32_en.md new file mode 100644 index 00000000000000..c6190e0c787814 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_nl32_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Tiny Cased model (from google) +author: John Snow Labs +name: t5_efficient_tiny_nl32 +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-tiny-nl32` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_nl32_en_4.3.0_3.0_1675123928277.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_nl32_en_4.3.0_3.0_1675123928277.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_tiny_nl32","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_tiny_nl32","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_tiny_nl32| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|145.4 MB| + +## References + +- https://huggingface.co/google/t5-efficient-tiny-nl32 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_nl6_en.md b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_nl6_en.md new file mode 100644 index 00000000000000..670b0cf048409b --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_nl6_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Tiny Cased model (from google) +author: John Snow Labs +name: t5_efficient_tiny_nl6 +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-tiny-nl6` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_nl6_en_4.3.0_3.0_1675123966207.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_nl6_en_4.3.0_3.0_1675123966207.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_tiny_nl6","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_tiny_nl6","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_tiny_nl6| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|53.2 MB| + +## References + +- https://huggingface.co/google/t5-efficient-tiny-nl6 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_nl8_en.md b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_nl8_en.md new file mode 100644 index 00000000000000..a3dfac31d03cc4 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_efficient_tiny_nl8_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Tiny Cased model (from google) +author: John Snow Labs +name: t5_efficient_tiny_nl8 +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-tiny-nl8` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_nl8_en_4.3.0_3.0_1675123995941.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_tiny_nl8_en_4.3.0_3.0_1675123995941.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_tiny_nl8","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_tiny_nl8","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_tiny_nl8| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|60.3 MB| + +## References + +- https://huggingface.co/google/t5-efficient-tiny-nl8 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_efficient_xl_nl2_en.md b/docs/_posts/Cabir40/2023-01-31-t5_efficient_xl_nl2_en.md new file mode 100644 index 00000000000000..88bb4f3739dae6 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_efficient_xl_nl2_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from google) +author: John Snow Labs +name: t5_efficient_xl_nl2 +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-xl-nl2` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_xl_nl2_en_4.3.0_3.0_1675124205513.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_xl_nl2_en_4.3.0_3.0_1675124205513.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_xl_nl2","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_xl_nl2","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_xl_nl2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|575.7 MB| + +## References + +- https://huggingface.co/google/t5-efficient-xl-nl2 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_efficient_xl_nl4_en.md b/docs/_posts/Cabir40/2023-01-31-t5_efficient_xl_nl4_en.md new file mode 100644 index 00000000000000..db7c7944cfab32 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_efficient_xl_nl4_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from google) +author: John Snow Labs +name: t5_efficient_xl_nl4 +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-efficient-xl-nl4` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_efficient_xl_nl4_en_4.3.0_3.0_1675124613893.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_efficient_xl_nl4_en_4.3.0_3.0_1675124613893.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_efficient_xl_nl4","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_efficient_xl_nl4","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_efficient_xl_nl4| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|1.0 GB| + +## References + +- https://huggingface.co/google/t5-efficient-xl-nl4 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/abs/2109.10686 +- https://arxiv.org/abs/2109.10686 +- https://github.com/google-research/google-research/issues/986#issuecomment-1035051145 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_finetuned_test_en.md b/docs/_posts/Cabir40/2023-01-31-t5_finetuned_test_en.md new file mode 100644 index 00000000000000..c814f6d12d3f8b --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_finetuned_test_en.md @@ -0,0 +1,85 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from osanseviero) +author: John Snow Labs +name: t5_finetuned_test +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-finetuned-test` is a English model originally trained by `osanseviero`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_finetuned_test_en_4.3.0_3.0_1675124670488.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_finetuned_test_en_4.3.0_3.0_1675124670488.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_finetuned_test","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_finetuned_test","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_finetuned_test| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|286.9 MB| + +## References + +- https://huggingface.co/osanseviero/t5-finetuned-test +- https://medium.com/@priya.dwivedi/fine-tuning-a-t5-transformer-for-any-summarization-task-82334c64c81 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_hotel_review_sentiment_tw.md b/docs/_posts/Cabir40/2023-01-31-t5_hotel_review_sentiment_tw.md new file mode 100644 index 00000000000000..3c2adf75faf297 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_hotel_review_sentiment_tw.md @@ -0,0 +1,84 @@ +--- +layout: model +title: Twi T5ForConditionalGeneration Cased model (from clhuang) +author: John Snow Labs +name: t5_hotel_review_sentiment +date: 2023-01-31 +tags: [tw, open_source, t5, tensorflow] +task: Text Generation +language: tw +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-hotel-review-sentiment` is a Twi model originally trained by `clhuang`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_hotel_review_sentiment_tw_4.3.0_3.0_1675124731516.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_hotel_review_sentiment_tw_4.3.0_3.0_1675124731516.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_hotel_review_sentiment","tw") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_hotel_review_sentiment","tw") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_hotel_review_sentiment| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|tw| +|Size:|1.0 GB| + +## References + +- https://huggingface.co/clhuang/t5-hotel-review-sentiment \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_informal_en.md b/docs/_posts/Cabir40/2023-01-31-t5_informal_en.md new file mode 100644 index 00000000000000..4491065a4fac1d --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_informal_en.md @@ -0,0 +1,85 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from SkolkovoInstitute) +author: John Snow Labs +name: t5_informal +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-informal` is a English model originally trained by `SkolkovoInstitute`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_informal_en_4.3.0_3.0_1675124817401.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_informal_en_4.3.0_3.0_1675124817401.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_informal","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_informal","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_informal| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|927.0 MB| + +## References + +- https://huggingface.co/SkolkovoInstitute/t5-informal +- https://aclanthology.org/N18-1012/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_inshorts_en.md b/docs/_posts/Cabir40/2023-01-31-t5_inshorts_en.md new file mode 100644 index 00000000000000..afabc46b328cb5 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_inshorts_en.md @@ -0,0 +1,89 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from lordtt13) +author: John Snow Labs +name: t5_inshorts +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-inshorts` is a English model originally trained by `lordtt13`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_inshorts_en_4.3.0_3.0_1675124897561.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_inshorts_en_4.3.0_3.0_1675124897561.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_inshorts","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_inshorts","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_inshorts| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|927.0 MB| + +## References + +- https://huggingface.co/lordtt13/t5-inshorts +- https://arxiv.org/abs/1910.10683 +- https://www.kaggle.com/shashichander009/inshorts-news-data +- https://github.com/lordtt13/transformers-experiments/blob/master/Custom%20Tasks/fine-tune-t5-for-summarization.ipynb +- https://github.com/lordtt13 +- https://www.linkedin.com/in/tanmay-thakur-6bb5a9154/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_measurement_time_en.md b/docs/_posts/Cabir40/2023-01-31-t5_measurement_time_en.md new file mode 100644 index 00000000000000..801dd3751be19d --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_measurement_time_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from dbernsohn) +author: John Snow Labs +name: t5_measurement_time +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5_measurement_time` is a English model originally trained by `dbernsohn`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_measurement_time_en_4.3.0_3.0_1675156797621.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_measurement_time_en_4.3.0_3.0_1675156797621.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_measurement_time","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_measurement_time","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_measurement_time| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|278.9 MB| + +## References + +- https://huggingface.co/dbernsohn/t5_measurement_time +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://www.tensorflow.org/datasets/catalog/math_dataset#mathdatasetmeasurement_time +- https://github.com/DorBernsohn/CodeLM/tree/main/MathLM +- https://www.linkedin.com/in/dor-bernsohn-70b2b1146/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_mini_nl8_fi.md b/docs/_posts/Cabir40/2023-01-31-t5_mini_nl8_fi.md new file mode 100644 index 00000000000000..44f885fd2ed8c1 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_mini_nl8_fi.md @@ -0,0 +1,100 @@ +--- +layout: model +title: Finnish T5ForConditionalGeneration Mini Cased model (from Finnish-NLP) +author: John Snow Labs +name: t5_mini_nl8 +date: 2023-01-31 +tags: [fi, open_source, t5, tensorflow] +task: Text Generation +language: fi +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-mini-nl8-finnish` is a Finnish model originally trained by `Finnish-NLP`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_mini_nl8_fi_4.3.0_3.0_1675124948833.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_mini_nl8_fi_4.3.0_3.0_1675124948833.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_mini_nl8","fi") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_mini_nl8","fi") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_mini_nl8| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|fi| +|Size:|315.9 MB| + +## References + +- https://huggingface.co/Finnish-NLP/t5-mini-nl8-finnish +- https://arxiv.org/abs/1910.10683 +- https://github.com/google-research/text-to-text-transfer-transformer +- https://github.com/google-research/text-to-text-transfer-transformer/blob/main/released_checkpoints.md#t511 +- https://arxiv.org/abs/2002.05202 +- https://arxiv.org/abs/2109.10686 +- http://urn.fi/urn:nbn:fi:lb-2017070501 +- http://urn.fi/urn:nbn:fi:lb-2021050401 +- http://urn.fi/urn:nbn:fi:lb-2018121001 +- http://urn.fi/urn:nbn:fi:lb-2020021803 +- https://sites.research.google/trc/about/ +- https://github.com/google-research/t5x +- https://github.com/spyysalo/yle-corpus +- https://github.com/aajanki/eduskunta-vkk +- https://sites.research.google/trc/ +- https://www.linkedin.com/in/aapotanskanen/ +- https://www.linkedin.com/in/rasmustoivanen/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_numbers_gcd_en.md b/docs/_posts/Cabir40/2023-01-31-t5_numbers_gcd_en.md new file mode 100644 index 00000000000000..8e9636e60aa968 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_numbers_gcd_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from dbernsohn) +author: John Snow Labs +name: t5_numbers_gcd +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5_numbers_gcd` is a English model originally trained by `dbernsohn`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_numbers_gcd_en_4.3.0_3.0_1675156829112.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_numbers_gcd_en_4.3.0_3.0_1675156829112.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_numbers_gcd","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_numbers_gcd","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_numbers_gcd| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|283.1 MB| + +## References + +- https://huggingface.co/dbernsohn/t5_numbers_gcd +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://www.tensorflow.org/datasets/catalog/math_dataset#mathdatasetnumbers_gcd +- https://github.com/DorBernsohn/CodeLM/tree/main/MathLM +- https://www.linkedin.com/in/dor-bernsohn-70b2b1146/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_paraphrase_paws_msrp_opinosis_en.md b/docs/_posts/Cabir40/2023-01-31-t5_paraphrase_paws_msrp_opinosis_en.md new file mode 100644 index 00000000000000..dd6ef82df6459e --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_paraphrase_paws_msrp_opinosis_en.md @@ -0,0 +1,85 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from ceshine) +author: John Snow Labs +name: t5_paraphrase_paws_msrp_opinosis +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-paraphrase-paws-msrp-opinosis` is a English model originally trained by `ceshine`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_paraphrase_paws_msrp_opinosis_en_4.3.0_3.0_1675125013995.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_paraphrase_paws_msrp_opinosis_en_4.3.0_3.0_1675125013995.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_paraphrase_paws_msrp_opinosis","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_paraphrase_paws_msrp_opinosis","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_paraphrase_paws_msrp_opinosis| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|923.5 MB| + +## References + +- https://huggingface.co/ceshine/t5-paraphrase-paws-msrp-opinosis +- https://github.com/ceshine/finetuning-t5/tree/master/paraphrase \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_paraphrase_quora_paws_en.md b/docs/_posts/Cabir40/2023-01-31-t5_paraphrase_quora_paws_en.md new file mode 100644 index 00000000000000..32c3cb8dd4e396 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_paraphrase_quora_paws_en.md @@ -0,0 +1,85 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from ceshine) +author: John Snow Labs +name: t5_paraphrase_quora_paws +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-paraphrase-quora-paws` is a English model originally trained by `ceshine`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_paraphrase_quora_paws_en_4.3.0_3.0_1675125097390.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_paraphrase_quora_paws_en_4.3.0_3.0_1675125097390.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_paraphrase_quora_paws","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_paraphrase_quora_paws","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_paraphrase_quora_paws| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|924.7 MB| + +## References + +- https://huggingface.co/ceshine/t5-paraphrase-quora-paws +- https://github.com/ceshine/finetuning-t5/tree/master/paraphrase \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_paraphraser_en.md b/docs/_posts/Cabir40/2023-01-31-t5_paraphraser_en.md new file mode 100644 index 00000000000000..b1437c4a9c8a84 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_paraphraser_en.md @@ -0,0 +1,84 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from valurank) +author: John Snow Labs +name: t5_paraphraser +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-paraphraser` is a English model originally trained by `valurank`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_paraphraser_en_4.3.0_3.0_1675125180653.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_paraphraser_en_4.3.0_3.0_1675125180653.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_paraphraser","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_paraphraser","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_paraphraser| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|920.7 MB| + +## References + +- https://huggingface.co/valurank/t5-paraphraser \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_podcast_summarisation_en.md b/docs/_posts/Cabir40/2023-01-31-t5_podcast_summarisation_en.md new file mode 100644 index 00000000000000..ce4c7b5227a105 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_podcast_summarisation_en.md @@ -0,0 +1,91 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from paulowoicho) +author: John Snow Labs +name: t5_podcast_summarisation +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-podcast-summarisation` is a English model originally trained by `paulowoicho`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_podcast_summarisation_en_4.3.0_3.0_1675125262437.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_podcast_summarisation_en_4.3.0_3.0_1675125262437.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_podcast_summarisation","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_podcast_summarisation","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_podcast_summarisation| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|927.4 MB| + +## References + +- https://huggingface.co/paulowoicho/t5-podcast-summarisation +- https://arxiv.org/abs/2004.04270 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/pdf/1910.10683.pdf +- https://arxiv.org/abs/2004.04270 +- https://github.com/paulowoicho/msc_project/blob/master/reformat.py +- https://github.com/abhimishra91/transformers-tutorials/blob/master/transformers_summarization_wandb.ipynb +- https://github.com/abhimishra91 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_punctuation_fr.md b/docs/_posts/Cabir40/2023-01-31-t5_punctuation_fr.md new file mode 100644 index 00000000000000..1da20293da5efe --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_punctuation_fr.md @@ -0,0 +1,85 @@ +--- +layout: model +title: French T5ForConditionalGeneration Cased model (from ZakaryaRouzki) +author: John Snow Labs +name: t5_punctuation +date: 2023-01-31 +tags: [fr, open_source, t5, tensorflow] +task: Text Generation +language: fr +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-punctuation` is a French model originally trained by `ZakaryaRouzki`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_punctuation_fr_4.3.0_3.0_1675125343642.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_punctuation_fr_4.3.0_3.0_1675125343642.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_punctuation","fr") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_punctuation","fr") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_punctuation| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|fr| +|Size:|925.1 MB| + +## References + +- https://huggingface.co/ZakaryaRouzki/t5-punctuation +- https://linkedin.com/in/rouzki \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_qa_squad2neg_en.md b/docs/_posts/Cabir40/2023-01-31-t5_qa_squad2neg_en.md new file mode 100644 index 00000000000000..82355157c31f58 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_qa_squad2neg_en.md @@ -0,0 +1,86 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from ThomasNLG) +author: John Snow Labs +name: t5_qa_squad2neg +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-qa_squad2neg-en` is a English model originally trained by `ThomasNLG`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_qa_squad2neg_en_4.3.0_3.0_1675125429554.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_qa_squad2neg_en_4.3.0_3.0_1675125429554.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_qa_squad2neg","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_qa_squad2neg","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_qa_squad2neg| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|922.0 MB| + +## References + +- https://huggingface.co/ThomasNLG/t5-qa_squad2neg-en +- https://github.com/ThomasScialom/QuestEval +- https://arxiv.org/abs/2103.12693 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_qa_webnlg_synth_en.md b/docs/_posts/Cabir40/2023-01-31-t5_qa_webnlg_synth_en.md new file mode 100644 index 00000000000000..14385a4ba0f075 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_qa_webnlg_synth_en.md @@ -0,0 +1,86 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from ThomasNLG) +author: John Snow Labs +name: t5_qa_webnlg_synth +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-qa_webnlg_synth-en` is a English model originally trained by `ThomasNLG`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_qa_webnlg_synth_en_4.3.0_3.0_1675125486836.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_qa_webnlg_synth_en_4.3.0_3.0_1675125486836.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_qa_webnlg_synth","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_qa_webnlg_synth","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_qa_webnlg_synth| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|271.7 MB| + +## References + +- https://huggingface.co/ThomasNLG/t5-qa_webnlg_synth-en +- https://github.com/ThomasScialom/QuestEval +- https://arxiv.org/abs/2104.07555 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_qg_squad1_en.md b/docs/_posts/Cabir40/2023-01-31-t5_qg_squad1_en.md new file mode 100644 index 00000000000000..eaa4922b31f0a4 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_qg_squad1_en.md @@ -0,0 +1,85 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from ThomasNLG) +author: John Snow Labs +name: t5_qg_squad1 +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-qg_squad1-en` is a English model originally trained by `ThomasNLG`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_qg_squad1_en_4.3.0_3.0_1675125547851.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_qg_squad1_en_4.3.0_3.0_1675125547851.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_qg_squad1","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_qg_squad1","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_qg_squad1| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|923.2 MB| + +## References + +- https://huggingface.co/ThomasNLG/t5-qg_squad1-en +- https://github.com/ThomasScialom/QuestEval \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_qg_webnlg_synth_en.md b/docs/_posts/Cabir40/2023-01-31-t5_qg_webnlg_synth_en.md new file mode 100644 index 00000000000000..8de1b7545cd04b --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_qg_webnlg_synth_en.md @@ -0,0 +1,86 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from ThomasNLG) +author: John Snow Labs +name: t5_qg_webnlg_synth +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-qg_webnlg_synth-en` is a English model originally trained by `ThomasNLG`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_qg_webnlg_synth_en_4.3.0_3.0_1675125600977.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_qg_webnlg_synth_en_4.3.0_3.0_1675125600977.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_qg_webnlg_synth","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_qg_webnlg_synth","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_qg_webnlg_synth| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|280.8 MB| + +## References + +- https://huggingface.co/ThomasNLG/t5-qg_webnlg_synth-en +- https://github.com/ThomasScialom/QuestEval +- https://arxiv.org/abs/2104.07555 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_qiita_title_generation_ja.md b/docs/_posts/Cabir40/2023-01-31-t5_qiita_title_generation_ja.md new file mode 100644 index 00000000000000..19b425d5a8b89b --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_qiita_title_generation_ja.md @@ -0,0 +1,85 @@ +--- +layout: model +title: Japanese T5ForConditionalGeneration Cased model (from sonoisa) +author: John Snow Labs +name: t5_qiita_title_generation +date: 2023-01-31 +tags: [ja, open_source, t5, tensorflow] +task: Text Generation +language: ja +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-qiita-title-generation` is a Japanese model originally trained by `sonoisa`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_qiita_title_generation_ja_4.3.0_3.0_1675125661321.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_qiita_title_generation_ja_4.3.0_3.0_1675125661321.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_qiita_title_generation","ja") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_qiita_title_generation","ja") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_qiita_title_generation| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|ja| +|Size:|925.3 MB| + +## References + +- https://huggingface.co/sonoisa/t5-qiita-title-generation +- https://qiita.com/sonoisa/items/30876467ad5a8a81821f \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_sl_small_sl.md b/docs/_posts/Cabir40/2023-01-31-t5_sl_small_sl.md new file mode 100644 index 00000000000000..aa2420e69bd588 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_sl_small_sl.md @@ -0,0 +1,85 @@ +--- +layout: model +title: Slovenian T5ForConditionalGeneration Small Cased model (from cjvt) +author: John Snow Labs +name: t5_sl_small +date: 2023-01-31 +tags: [sl, open_source, t5, tensorflow] +task: Text Generation +language: sl +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-sl-small` is a Slovenian model originally trained by `cjvt`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_sl_small_sl_4.3.0_3.0_1675125783776.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_sl_small_sl_4.3.0_3.0_1675125783776.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_sl_small","sl") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_sl_small","sl") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_sl_small| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|sl| +|Size:|178.8 MB| + +## References + +- https://huggingface.co/cjvt/t5-sl-small +- https://arxiv.org/abs/2207.13988 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_small_bahasa_cased_ms.md b/docs/_posts/Cabir40/2023-01-31-t5_small_bahasa_cased_ms.md new file mode 100644 index 00000000000000..37f9275f41040b --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_small_bahasa_cased_ms.md @@ -0,0 +1,87 @@ +--- +layout: model +title: Malay T5ForConditionalGeneration Small Cased model (from mesolitica) +author: John Snow Labs +name: t5_small_bahasa_cased +date: 2023-01-31 +tags: [ms, open_source, t5, tensorflow] +task: Text Generation +language: ms +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-small-bahasa-cased` is a Malay model originally trained by `mesolitica`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_small_bahasa_cased_ms_4.3.0_3.0_1675125885354.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_small_bahasa_cased_ms_4.3.0_3.0_1675125885354.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_small_bahasa_cased","ms") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_small_bahasa_cased","ms") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_small_bahasa_cased| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|ms| +|Size:|147.7 MB| + +## References + +- https://huggingface.co/mesolitica/t5-small-bahasa-cased +- https://github.com/huseinzol05/malaya/tree/master/pretrained-model/t5/prepare +- https://github.com/google-research/text-to-text-transfer-transformer +- https://github.com/huseinzol05/Malaya/tree/master/pretrained-model/t5 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_small_bashsql_en.md b/docs/_posts/Cabir40/2023-01-31-t5_small_bashsql_en.md new file mode 100644 index 00000000000000..e93ec80c411370 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_small_bashsql_en.md @@ -0,0 +1,87 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from NeuML) +author: John Snow Labs +name: t5_small_bashsql +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-small-bashsql` is a English model originally trained by `NeuML`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_small_bashsql_en_4.3.0_3.0_1675125918159.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_small_bashsql_en_4.3.0_3.0_1675125918159.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_small_bashsql","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_small_bashsql","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_small_bashsql| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|260.9 MB| + +## References + +- https://huggingface.co/NeuML/t5-small-bashsql +- https://github.com/neuml/txtai +- https://en.wikipedia.org/wiki/Bash_(Unix_shell) +- https://github.com/neuml/txtai/tree/master/models/bashsql \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_small_en.md b/docs/_posts/Cabir40/2023-01-31-t5_small_en.md new file mode 100644 index 00000000000000..f26578807426b1 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_small_en.md @@ -0,0 +1,96 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from khanglam7012) +author: John Snow Labs +name: t5_small +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-small` is a English model originally trained by `khanglam7012`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_small_en_4.3.0_3.0_1675125819094.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_small_en_4.3.0_3.0_1675125819094.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_small","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_small","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_small| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|253.6 MB| + +## References + +- https://huggingface.co/khanglam7012/t5-small +- https://user-images.githubusercontent.com/49101362/116334480-f5e57a00-a7dd-11eb-987c-186477f94b6e.png +- https://pypi.org/project/keytotext/ +- https://pepy.tech/project/keytotext +- https://colab.research.google.com/github/gagan3012/keytotext/blob/master/Examples/K2T.ipynb +- https://share.streamlit.io/gagan3012/keytotext/UI/app.py +- https://github.com/gagan3012/keytotext/tree/master/Training%20Notebooks +- https://colab.research.google.com/github/gagan3012/keytotext/blob/master/Examples/K2T.ipynb +- https://github.com/gagan3012/keytotext/tree/master/Examples +- https://user-images.githubusercontent.com/49101362/116220679-90e64180-a755-11eb-9246-82d93d924a6c.png +- https://share.streamlit.io/gagan3012/keytotext/UI/app.py +- https://github.com/gagan3012/streamlit-tags +- https://user-images.githubusercontent.com/49101362/116162205-fc042980-a6fd-11eb-892e-8f6902f193f4.png \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_small_finetuned_bias_en.md b/docs/_posts/Cabir40/2023-01-31-t5_small_finetuned_bias_en.md new file mode 100644 index 00000000000000..867874c43cd47c --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_small_finetuned_bias_en.md @@ -0,0 +1,85 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from erickfm) +author: John Snow Labs +name: t5_small_finetuned_bias +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-small-finetuned-bias` is a English model originally trained by `erickfm`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_small_finetuned_bias_en_4.3.0_3.0_1675125947945.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_small_finetuned_bias_en_4.3.0_3.0_1675125947945.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_small_finetuned_bias","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_small_finetuned_bias","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_small_finetuned_bias| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|284.4 MB| + +## References + +- https://huggingface.co/erickfm/t5-small-finetuned-bias +- https://github.com/rpryzant/neutralizing-bias \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_small_finetuned_emotion_en.md b/docs/_posts/Cabir40/2023-01-31-t5_small_finetuned_emotion_en.md new file mode 100644 index 00000000000000..5737f758058535 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_small_finetuned_emotion_en.md @@ -0,0 +1,95 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from mrm8488) +author: John Snow Labs +name: t5_small_finetuned_emotion +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-small-finetuned-emotion` is a English model originally trained by `mrm8488`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_small_finetuned_emotion_en_4.3.0_3.0_1675125983774.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_small_finetuned_emotion_en_4.3.0_3.0_1675125983774.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_small_finetuned_emotion","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_small_finetuned_emotion","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_small_finetuned_emotion| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|263.3 MB| + +## References + +- https://huggingface.co/mrm8488/t5-small-finetuned-emotion +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://github.com/dair-ai/emotion_dataset +- https://arxiv.org/pdf/1910.10683.pdf +- https://i.imgur.com/jVFMMWR.png +- https://twitter.com/omarsar0 +- https://github.com/dair-ai/emotion_dataset +- https://github.com/patil-suraj/exploring-T5/blob/master/t5_fine_tuning.ipynb +- https://github.com/patil-suraj +- https://i.imgur.com/JBtAwPx.png +- https://twitter.com/mrm8488 +- https://www.linkedin.com/in/manuel-romero-cs/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_small_finetuned_imdb_sentiment_en.md b/docs/_posts/Cabir40/2023-01-31-t5_small_finetuned_imdb_sentiment_en.md new file mode 100644 index 00000000000000..cc6761092c3332 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_small_finetuned_imdb_sentiment_en.md @@ -0,0 +1,91 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from mrm8488) +author: John Snow Labs +name: t5_small_finetuned_imdb_sentiment +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-small-finetuned-imdb-sentiment` is a English model originally trained by `mrm8488`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_small_finetuned_imdb_sentiment_en_4.3.0_3.0_1675126033560.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_small_finetuned_imdb_sentiment_en_4.3.0_3.0_1675126033560.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_small_finetuned_imdb_sentiment","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_small_finetuned_imdb_sentiment","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_small_finetuned_imdb_sentiment| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|267.7 MB| + +## References + +- https://huggingface.co/mrm8488/t5-small-finetuned-imdb-sentiment +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/pdf/1910.10683.pdf +- https://camo.githubusercontent.com/623b4dea0b653f2ad3f36c71ebfe749a677ac0a1/68747470733a2f2f6d69726f2e6d656469756d2e636f6d2f6d61782f343030362f312a44304a31674e51663876727255704b657944387750412e706e67 +- https://github.com/patil-suraj/exploring-T5/blob/master/t5_fine_tuning.ipynb +- https://github.com/patil-suraj +- https://twitter.com/mrm8488 +- https://www.linkedin.com/in/manuel-romero-cs/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_small_finetuned_quora_for_paraphrasing_en.md b/docs/_posts/Cabir40/2023-01-31-t5_small_finetuned_quora_for_paraphrasing_en.md new file mode 100644 index 00000000000000..a7439644844e66 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_small_finetuned_quora_for_paraphrasing_en.md @@ -0,0 +1,90 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from mrm8488) +author: John Snow Labs +name: t5_small_finetuned_quora_for_paraphrasing +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-small-finetuned-quora-for-paraphrasing` is a English model originally trained by `mrm8488`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_small_finetuned_quora_for_paraphrasing_en_4.3.0_3.0_1675126066647.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_small_finetuned_quora_for_paraphrasing_en_4.3.0_3.0_1675126066647.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_small_finetuned_quora_for_paraphrasing","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_small_finetuned_quora_for_paraphrasing","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_small_finetuned_quora_for_paraphrasing| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|269.4 MB| + +## References + +- https://huggingface.co/mrm8488/t5-small-finetuned-quora-for-paraphrasing +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/pdf/1910.10683.pdf +- https://i.imgur.com/jVFMMWR.png +- https://colab.research.google.com/github/patil-suraj/exploring-T5/blob/master/T5_on_TPU.ipynb +- https://twitter.com/mrm8488 +- https://www.linkedin.com/in/manuel-romero-cs/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_small_finetuned_spanish_to_quechua_xx.md b/docs/_posts/Cabir40/2023-01-31-t5_small_finetuned_spanish_to_quechua_xx.md new file mode 100644 index 00000000000000..4cfc589698f06e --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_small_finetuned_spanish_to_quechua_xx.md @@ -0,0 +1,84 @@ +--- +layout: model +title: Multilingual T5ForConditionalGeneration Small Cased model (from hackathon-pln-es) +author: John Snow Labs +name: t5_small_finetuned_spanish_to_quechua +date: 2023-01-31 +tags: [qu, es, open_source, t5, xx, tensorflow] +task: Text Generation +language: xx +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-small-finetuned-spanish-to-quechua` is a Multilingual model originally trained by `hackathon-pln-es`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_small_finetuned_spanish_to_quechua_xx_4.3.0_3.0_1675126096661.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_small_finetuned_spanish_to_quechua_xx_4.3.0_3.0_1675126096661.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_small_finetuned_spanish_to_quechua","xx") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_small_finetuned_spanish_to_quechua","xx") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_small_finetuned_spanish_to_quechua| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|xx| +|Size:|282.6 MB| + +## References + +- https://huggingface.co/hackathon-pln-es/t5-small-finetuned-spanish-to-quechua \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_small_finetuned_squadv1_en.md b/docs/_posts/Cabir40/2023-01-31-t5_small_finetuned_squadv1_en.md new file mode 100644 index 00000000000000..ee1fbf54c32a2e --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_small_finetuned_squadv1_en.md @@ -0,0 +1,92 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from mrm8488) +author: John Snow Labs +name: t5_small_finetuned_squadv1 +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-small-finetuned-squadv1` is a English model originally trained by `mrm8488`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_small_finetuned_squadv1_en_4.3.0_3.0_1675126129461.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_small_finetuned_squadv1_en_4.3.0_3.0_1675126129461.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_small_finetuned_squadv1","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_small_finetuned_squadv1","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_small_finetuned_squadv1| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|274.6 MB| + +## References + +- https://huggingface.co/mrm8488/t5-small-finetuned-squadv1 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://rajpurkar.github.io/SQuAD-explorer/ +- https://arxiv.org/pdf/1910.10683.pdf +- https://i.imgur.com/jVFMMWR.png +- https://colab.research.google.com/github/patil-suraj/exploring-T5/blob/master/T5_on_TPU.ipynb +- https://twitter.com/psuraj28 +- https://twitter.com/mrm8488 +- https://www.linkedin.com/in/manuel-romero-cs/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_small_finetuned_squadv2_en.md b/docs/_posts/Cabir40/2023-01-31-t5_small_finetuned_squadv2_en.md new file mode 100644 index 00000000000000..8f3dfb8041d804 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_small_finetuned_squadv2_en.md @@ -0,0 +1,92 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from mrm8488) +author: John Snow Labs +name: t5_small_finetuned_squadv2 +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-small-finetuned-squadv2` is a English model originally trained by `mrm8488`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_small_finetuned_squadv2_en_4.3.0_3.0_1675126161506.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_small_finetuned_squadv2_en_4.3.0_3.0_1675126161506.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_small_finetuned_squadv2","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_small_finetuned_squadv2","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_small_finetuned_squadv2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|275.3 MB| + +## References + +- https://huggingface.co/mrm8488/t5-small-finetuned-squadv2 +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://rajpurkar.github.io/SQuAD-explorer/ +- https://arxiv.org/pdf/1910.10683.pdf +- https://i.imgur.com/jVFMMWR.png +- https://colab.research.google.com/github/patil-suraj/exploring-T5/blob/master/T5_on_TPU.ipynb +- https://twitter.com/psuraj28 +- https://twitter.com/mrm8488 +- https://www.linkedin.com/in/manuel-romero-cs/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_small_finetuned_text2log_en.md b/docs/_posts/Cabir40/2023-01-31-t5_small_finetuned_text2log_en.md new file mode 100644 index 00000000000000..01157427620ea0 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_small_finetuned_text2log_en.md @@ -0,0 +1,84 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from mrm8488) +author: John Snow Labs +name: t5_small_finetuned_text2log +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-small-finetuned-text2log` is a English model originally trained by `mrm8488`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_small_finetuned_text2log_en_4.3.0_3.0_1675126191501.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_small_finetuned_text2log_en_4.3.0_3.0_1675126191501.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_small_finetuned_text2log","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_small_finetuned_text2log","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_small_finetuned_text2log| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|286.3 MB| + +## References + +- https://huggingface.co/mrm8488/t5-small-finetuned-text2log \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_small_finetuned_wikisql_en.md b/docs/_posts/Cabir40/2023-01-31-t5_small_finetuned_wikisql_en.md new file mode 100644 index 00000000000000..cd863b38f34b8c --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_small_finetuned_wikisql_en.md @@ -0,0 +1,92 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from mrm8488) +author: John Snow Labs +name: t5_small_finetuned_wikisql +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-small-finetuned-wikiSQL` is a English model originally trained by `mrm8488`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_small_finetuned_wikisql_en_4.3.0_3.0_1675126227801.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_small_finetuned_wikisql_en_4.3.0_3.0_1675126227801.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_small_finetuned_wikisql","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_small_finetuned_wikisql","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_small_finetuned_wikisql| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|262.1 MB| + +## References + +- https://huggingface.co/mrm8488/t5-small-finetuned-wikiSQL +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://github.com/salesforce/WikiSQL +- https://arxiv.org/pdf/1910.10683.pdf +- https://i.imgur.com/jVFMMWR.png +- https://github.com/patil-suraj/exploring-T5/blob/master/t5_fine_tuning.ipynb +- https://github.com/patil-suraj +- https://twitter.com/mrm8488 +- https://www.linkedin.com/in/manuel-romero-cs/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_small_german_de.md b/docs/_posts/Cabir40/2023-01-31-t5_small_german_de.md new file mode 100644 index 00000000000000..4faa2d4e549615 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_small_german_de.md @@ -0,0 +1,85 @@ +--- +layout: model +title: German T5ForConditionalGeneration Small Cased model (from Shahm) +author: John Snow Labs +name: t5_small_german +date: 2023-01-31 +tags: [de, open_source, t5, tensorflow] +task: Text Generation +language: de +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-small-german` is a German model originally trained by `Shahm`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_small_german_de_4.3.0_3.0_1675126257283.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_small_german_de_4.3.0_3.0_1675126257283.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_small_german","de") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_small_german","de") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_small_german| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|de| +|Size:|285.5 MB| + +## References + +- https://huggingface.co/Shahm/t5-small-german +- https://paperswithcode.com/sota?task=Summarization&dataset=mlsum+de \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_small_grammar_correction_de.md b/docs/_posts/Cabir40/2023-01-31-t5_small_grammar_correction_de.md new file mode 100644 index 00000000000000..acd4183030da5b --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_small_grammar_correction_de.md @@ -0,0 +1,87 @@ +--- +layout: model +title: German T5ForConditionalGeneration Small Cased model (from aiassociates) +author: John Snow Labs +name: t5_small_grammar_correction +date: 2023-01-31 +tags: [de, open_source, t5, tensorflow] +task: Text Generation +language: de +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-small-grammar-correction-german` is a German model originally trained by `aiassociates`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_small_grammar_correction_de_4.3.0_3.0_1675126287089.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_small_grammar_correction_de_4.3.0_3.0_1675126287089.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_small_grammar_correction","de") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_small_grammar_correction","de") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_small_grammar_correction| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|de| +|Size:|288.7 MB| + +## References + +- https://huggingface.co/aiassociates/t5-small-grammar-correction-german +- https://github.com/EricFillion/happy-transformer +- https://www.ai.associates/ +- https://www.linkedin.com/company/ai-associates \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_small_grammar_ro.md b/docs/_posts/Cabir40/2023-01-31-t5_small_grammar_ro.md new file mode 100644 index 00000000000000..d26f5a6ad39c18 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_small_grammar_ro.md @@ -0,0 +1,85 @@ +--- +layout: model +title: Romanian T5ForConditionalGeneration Small Cased model (from BlackKakapo) +author: John Snow Labs +name: t5_small_grammar +date: 2023-01-31 +tags: [ro, open_source, t5, tensorflow] +task: Text Generation +language: ro +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-small-grammar-ro` is a Romanian model originally trained by `BlackKakapo`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_small_grammar_ro_4.3.0_3.0_1675126317557.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_small_grammar_ro_4.3.0_3.0_1675126317557.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_small_grammar","ro") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_small_grammar","ro") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_small_grammar| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|ro| +|Size:|284.0 MB| + +## References + +- https://huggingface.co/BlackKakapo/t5-small-grammar-ro +- https://img.shields.io/badge/V.1-03.08.2022-brightgreen \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_small_grammar_v2_ro.md b/docs/_posts/Cabir40/2023-01-31-t5_small_grammar_v2_ro.md new file mode 100644 index 00000000000000..c5261b27699930 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_small_grammar_v2_ro.md @@ -0,0 +1,85 @@ +--- +layout: model +title: Romanian T5ForConditionalGeneration Small Cased model (from BlackKakapo) +author: John Snow Labs +name: t5_small_grammar_v2 +date: 2023-01-31 +tags: [ro, open_source, t5, tensorflow] +task: Text Generation +language: ro +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-small-grammar-ro-v2` is a Romanian model originally trained by `BlackKakapo`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_small_grammar_v2_ro_4.3.0_3.0_1675126347481.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_small_grammar_v2_ro_4.3.0_3.0_1675126347481.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_small_grammar_v2","ro") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_small_grammar_v2","ro") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_small_grammar_v2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|ro| +|Size:|288.1 MB| + +## References + +- https://huggingface.co/BlackKakapo/t5-small-grammar-ro-v2 +- https://img.shields.io/badge/V.2-06.08.2022-brightgreen \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_small_headline_generator_en.md b/docs/_posts/Cabir40/2023-01-31-t5_small_headline_generator_en.md new file mode 100644 index 00000000000000..aae6495911fca8 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_small_headline_generator_en.md @@ -0,0 +1,84 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from JulesBelveze) +author: John Snow Labs +name: t5_small_headline_generator +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-small-headline-generator` is a English model originally trained by `JulesBelveze`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_small_headline_generator_en_4.3.0_3.0_1675126378182.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_small_headline_generator_en_4.3.0_3.0_1675126378182.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_small_headline_generator","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_small_headline_generator","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_small_headline_generator| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|282.9 MB| + +## References + +- https://huggingface.co/JulesBelveze/t5-small-headline-generator \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_small_lm_adapt_en.md b/docs/_posts/Cabir40/2023-01-31-t5_small_lm_adapt_en.md new file mode 100644 index 00000000000000..30ac1167f1becb --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_small_lm_adapt_en.md @@ -0,0 +1,90 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_small_lm_adapt +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-small-lm-adapt` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_small_lm_adapt_en_4.3.0_3.0_1675126486527.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_small_lm_adapt_en_4.3.0_3.0_1675126486527.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_small_lm_adapt","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_small_lm_adapt","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_small_lm_adapt| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|179.3 MB| + +## References + +- https://huggingface.co/google/t5-small-lm-adapt +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://github.com/google-research/text-to-text-transfer-transformer/blob/main/released_checkpoints.md#lm-adapted-t511lm100k +- https://arxiv.org/abs/2002.05202 +- https://arxiv.org/pdf/1910.10683.pdf +- https://arxiv.org/pdf/1910.10683.pdf +- https://camo.githubusercontent.com/623b4dea0b653f2ad3f36c71ebfe749a677ac0a1/68747470733a2f2f6d69726f2e6d656469756d2e636f6d2f6d61782f343030362f312a44304a31674e51663876727255704b657944387750412e706e67 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_small_ncc_lm_xx.md b/docs/_posts/Cabir40/2023-01-31-t5_small_ncc_lm_xx.md new file mode 100644 index 00000000000000..d2f78ba27cbf48 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_small_ncc_lm_xx.md @@ -0,0 +1,90 @@ +--- +layout: model +title: Multilingual T5ForConditionalGeneration Small Cased model (from north) +author: John Snow Labs +name: t5_small_ncc_lm +date: 2023-01-31 +tags: [is, nn, en, "no", sv, open_source, t5, xx, tensorflow] +task: Text Generation +language: xx +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5_small_NCC_lm` is a Multilingual model originally trained by `north`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_small_ncc_lm_xx_4.3.0_3.0_1675157076420.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_small_ncc_lm_xx_4.3.0_3.0_1675157076420.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_small_ncc_lm","xx") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_small_ncc_lm","xx") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_small_ncc_lm| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|xx| +|Size:|1.6 GB| + +## References + +- https://huggingface.co/north/t5_small_NCC_lm +- https://github.com/google-research/text-to-text-transfer-transformer +- https://github.com/google-research/t5x +- https://arxiv.org/abs/2104.09617 +- https://arxiv.org/abs/2104.09617 +- https://arxiv.org/pdf/1910.10683.pdf +- https://sites.research.google/trc/about/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_small_ncc_xx.md b/docs/_posts/Cabir40/2023-01-31-t5_small_ncc_xx.md new file mode 100644 index 00000000000000..23b6c0eabeeae4 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_small_ncc_xx.md @@ -0,0 +1,90 @@ +--- +layout: model +title: Multilingual T5ForConditionalGeneration Small Cased model (from north) +author: John Snow Labs +name: t5_small_ncc +date: 2023-01-31 +tags: [is, nn, en, "no", sv, open_source, t5, xx, tensorflow] +task: Text Generation +language: xx +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5_small_NCC` is a Multilingual model originally trained by `north`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_small_ncc_xx_4.3.0_3.0_1675156926295.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_small_ncc_xx_4.3.0_3.0_1675156926295.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_small_ncc","xx") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_small_ncc","xx") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_small_ncc| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|xx| +|Size:|1.6 GB| + +## References + +- https://huggingface.co/north/t5_small_NCC +- https://github.com/google-research/text-to-text-transfer-transformer +- https://github.com/google-research/t5x +- https://arxiv.org/abs/2104.09617 +- https://arxiv.org/abs/2104.09617 +- https://arxiv.org/pdf/1910.10683.pdf +- https://sites.research.google/trc/about/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_small_next_word_generator_qoogle_en.md b/docs/_posts/Cabir40/2023-01-31-t5_small_next_word_generator_qoogle_en.md new file mode 100644 index 00000000000000..6c999c1c4a7bd8 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_small_next_word_generator_qoogle_en.md @@ -0,0 +1,84 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from allenai) +author: John Snow Labs +name: t5_small_next_word_generator_qoogle +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-small-next-word-generator-qoogle` is a English model originally trained by `allenai`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_small_next_word_generator_qoogle_en_4.3.0_3.0_1675126551905.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_small_next_word_generator_qoogle_en_4.3.0_3.0_1675126551905.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_small_next_word_generator_qoogle","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_small_next_word_generator_qoogle","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_small_next_word_generator_qoogle| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|148.1 MB| + +## References + +- https://huggingface.co/allenai/t5-small-next-word-generator-qoogle \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_small_nl16_fi.md b/docs/_posts/Cabir40/2023-01-31-t5_small_nl16_fi.md new file mode 100644 index 00000000000000..6600a619f9ccfd --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_small_nl16_fi.md @@ -0,0 +1,100 @@ +--- +layout: model +title: Finnish T5ForConditionalGeneration Small Cased model (from Finnish-NLP) +author: John Snow Labs +name: t5_small_nl16 +date: 2023-01-31 +tags: [fi, open_source, t5, tensorflow] +task: Text Generation +language: fi +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-small-nl16-finnish` is a Finnish model originally trained by `Finnish-NLP`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_small_nl16_fi_4.3.0_3.0_1675126599389.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_small_nl16_fi_4.3.0_3.0_1675126599389.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_small_nl16","fi") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_small_nl16","fi") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_small_nl16| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|fi| +|Size:|751.2 MB| + +## References + +- https://huggingface.co/Finnish-NLP/t5-small-nl16-finnish +- https://arxiv.org/abs/1910.10683 +- https://github.com/google-research/text-to-text-transfer-transformer +- https://github.com/google-research/text-to-text-transfer-transformer/blob/main/released_checkpoints.md#t511 +- https://arxiv.org/abs/2002.05202 +- https://arxiv.org/abs/2109.10686 +- http://urn.fi/urn:nbn:fi:lb-2017070501 +- http://urn.fi/urn:nbn:fi:lb-2021050401 +- http://urn.fi/urn:nbn:fi:lb-2018121001 +- http://urn.fi/urn:nbn:fi:lb-2020021803 +- https://sites.research.google/trc/about/ +- https://github.com/google-research/t5x +- https://github.com/spyysalo/yle-corpus +- https://github.com/aajanki/eduskunta-vkk +- https://sites.research.google/trc/ +- https://www.linkedin.com/in/aapotanskanen/ +- https://www.linkedin.com/in/rasmustoivanen/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_small_paraphrase_ro.md b/docs/_posts/Cabir40/2023-01-31-t5_small_paraphrase_ro.md new file mode 100644 index 00000000000000..f6331f30983fc9 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_small_paraphrase_ro.md @@ -0,0 +1,85 @@ +--- +layout: model +title: Romanian T5ForConditionalGeneration Small Cased model (from BlackKakapo) +author: John Snow Labs +name: t5_small_paraphrase +date: 2023-01-31 +tags: [ro, open_source, t5, tensorflow] +task: Text Generation +language: ro +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-small-paraphrase-ro` is a Romanian model originally trained by `BlackKakapo`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_small_paraphrase_ro_4.3.0_3.0_1675126644590.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_small_paraphrase_ro_4.3.0_3.0_1675126644590.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_small_paraphrase","ro") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_small_paraphrase","ro") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_small_paraphrase| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|ro| +|Size:|288.6 MB| + +## References + +- https://huggingface.co/BlackKakapo/t5-small-paraphrase-ro +- https://img.shields.io/badge/V.1-03.08.2022-brightgreen \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_small_paraphrase_v2_ro.md b/docs/_posts/Cabir40/2023-01-31-t5_small_paraphrase_v2_ro.md new file mode 100644 index 00000000000000..d46addad33b463 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_small_paraphrase_v2_ro.md @@ -0,0 +1,85 @@ +--- +layout: model +title: Romanian T5ForConditionalGeneration Small Cased model (from BlackKakapo) +author: John Snow Labs +name: t5_small_paraphrase_v2 +date: 2023-01-31 +tags: [ro, open_source, t5, tensorflow] +task: Text Generation +language: ro +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-small-paraphrase-ro-v2` is a Romanian model originally trained by `BlackKakapo`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_small_paraphrase_v2_ro_4.3.0_3.0_1675155540424.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_small_paraphrase_v2_ro_4.3.0_3.0_1675155540424.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_small_paraphrase_v2","ro") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_small_paraphrase_v2","ro") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_small_paraphrase_v2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|ro| +|Size:|288.8 MB| + +## References + +- https://huggingface.co/BlackKakapo/t5-small-paraphrase-ro-v2 +- https://img.shields.io/badge/V.2-17.08.2022-brightgreen \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_small_quora_en.md b/docs/_posts/Cabir40/2023-01-31-t5_small_quora_en.md new file mode 100644 index 00000000000000..c7edb333baeb10 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_small_quora_en.md @@ -0,0 +1,86 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from hetpandya) +author: John Snow Labs +name: t5_small_quora +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-small-quora` is a English model originally trained by `hetpandya`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_small_quora_en_4.3.0_3.0_1675155570316.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_small_quora_en_4.3.0_3.0_1675155570316.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_small_quora","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_small_quora","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_small_quora| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|288.7 MB| + +## References + +- https://huggingface.co/hetpandya/t5-small-quora +- https://github.com/hetpandya +- https://www.linkedin.com/in/het-pandya \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_small_squad11_en.md b/docs/_posts/Cabir40/2023-01-31-t5_small_squad11_en.md new file mode 100644 index 00000000000000..1b5b8a529088d2 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_small_squad11_en.md @@ -0,0 +1,84 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from allenai) +author: John Snow Labs +name: t5_small_squad11 +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-small-squad11` is a English model originally trained by `allenai`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_small_squad11_en_4.3.0_3.0_1675155640438.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_small_squad11_en_4.3.0_3.0_1675155640438.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_small_squad11","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_small_squad11","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_small_squad11| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|148.2 MB| + +## References + +- https://huggingface.co/allenai/t5-small-squad11 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_small_squad2_next_word_generator_squad_en.md b/docs/_posts/Cabir40/2023-01-31-t5_small_squad2_next_word_generator_squad_en.md new file mode 100644 index 00000000000000..75df828da977ae --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_small_squad2_next_word_generator_squad_en.md @@ -0,0 +1,84 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from allenai) +author: John Snow Labs +name: t5_small_squad2_next_word_generator_squad +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-small-squad2-next-word-generator-squad` is a English model originally trained by `allenai`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_small_squad2_next_word_generator_squad_en_4.3.0_3.0_1675155704406.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_small_squad2_next_word_generator_squad_en_4.3.0_3.0_1675155704406.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_small_squad2_next_word_generator_squad","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_small_squad2_next_word_generator_squad","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_small_squad2_next_word_generator_squad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|148.1 MB| + +## References + +- https://huggingface.co/allenai/t5-small-squad2-next-word-generator-squad \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_small_squad2_question_generation_en.md b/docs/_posts/Cabir40/2023-01-31-t5_small_squad2_question_generation_en.md new file mode 100644 index 00000000000000..ec8158dcccfc0d --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_small_squad2_question_generation_en.md @@ -0,0 +1,84 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from allenai) +author: John Snow Labs +name: t5_small_squad2_question_generation +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-small-squad2-question-generation` is a English model originally trained by `allenai`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_small_squad2_question_generation_en_4.3.0_3.0_1675155768128.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_small_squad2_question_generation_en_4.3.0_3.0_1675155768128.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_small_squad2_question_generation","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_small_squad2_question_generation","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_small_squad2_question_generation| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|148.2 MB| + +## References + +- https://huggingface.co/allenai/t5-small-squad2-question-generation \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_small_ssm_en.md b/docs/_posts/Cabir40/2023-01-31-t5_small_ssm_en.md new file mode 100644 index 00000000000000..0f75c5db9e6e7e --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_small_ssm_en.md @@ -0,0 +1,89 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_small_ssm +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-small-ssm` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_small_ssm_en_4.3.0_3.0_1675155844003.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_small_ssm_en_4.3.0_3.0_1675155844003.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_small_ssm","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_small_ssm","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_small_ssm| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|179.4 MB| + +## References + +- https://huggingface.co/google/t5-small-ssm +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/pdf/2002.08909.pdf +- https://arxiv.org/abs/1910.10683.pdf +- https://goo.gle/t5-cbqa +- https://raw.githubusercontent.com/patrickvonplaten/scientific_images/master/how_much_know_ledge_image.png \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_small_ssm_nq_en.md b/docs/_posts/Cabir40/2023-01-31-t5_small_ssm_nq_en.md new file mode 100644 index 00000000000000..aaf6f59dea44cd --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_small_ssm_nq_en.md @@ -0,0 +1,89 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_small_ssm_nq +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-small-ssm-nq` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_small_ssm_nq_en_4.3.0_3.0_1675155920334.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_small_ssm_nq_en_4.3.0_3.0_1675155920334.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_small_ssm_nq","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_small_ssm_nq","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_small_ssm_nq| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|179.4 MB| + +## References + +- https://huggingface.co/google/t5-small-ssm-nq +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://arxiv.org/pdf/2002.08909.pdf +- https://arxiv.org/abs/1910.10683.pdf +- https://goo.gle/t5-cbqa +- https://raw.githubusercontent.com/patrickvonplaten/scientific_images/master/how_much_know_ledge_image.png \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_small_summarization_cased_id.md b/docs/_posts/Cabir40/2023-01-31-t5_small_summarization_cased_id.md new file mode 100644 index 00000000000000..49af2547321c02 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_small_summarization_cased_id.md @@ -0,0 +1,85 @@ +--- +layout: model +title: Indonesian T5ForConditionalGeneration Small Cased model (from panggi) +author: John Snow Labs +name: t5_small_summarization_cased +date: 2023-01-31 +tags: [id, open_source, t5, tensorflow] +task: Text Generation +language: id +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-small-indonesian-summarization-cased` is a Indonesian model originally trained by `panggi`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_small_summarization_cased_id_4.3.0_3.0_1675126407873.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_small_summarization_cased_id_4.3.0_3.0_1675126407873.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_small_summarization_cased","id") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_small_summarization_cased","id") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_small_summarization_cased| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|id| +|Size:|287.9 MB| + +## References + +- https://huggingface.co/panggi/t5-small-indonesian-summarization-cased +- https://github.com/kata-ai/indosum \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_small_summarization_ro.md b/docs/_posts/Cabir40/2023-01-31-t5_small_summarization_ro.md new file mode 100644 index 00000000000000..0267f593c0c236 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_small_summarization_ro.md @@ -0,0 +1,85 @@ +--- +layout: model +title: Romanian T5ForConditionalGeneration Small Cased model (from BlackKakapo) +author: John Snow Labs +name: t5_small_summarization +date: 2023-01-31 +tags: [ro, open_source, t5, tensorflow] +task: Text Generation +language: ro +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-small-summarization-ro` is a Romanian model originally trained by `BlackKakapo`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_small_summarization_ro_4.3.0_3.0_1675155949366.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_small_summarization_ro_4.3.0_3.0_1675155949366.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_small_summarization","ro") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_small_summarization","ro") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_small_summarization| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|ro| +|Size:|287.1 MB| + +## References + +- https://huggingface.co/BlackKakapo/t5-small-summarization-ro +- https://img.shields.io/badge/V.1-18.10.2022-brightgreen \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_small_tapaco_en.md b/docs/_posts/Cabir40/2023-01-31-t5_small_tapaco_en.md new file mode 100644 index 00000000000000..951b2acd3607ff --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_small_tapaco_en.md @@ -0,0 +1,87 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from hetpandya) +author: John Snow Labs +name: t5_small_tapaco +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-small-tapaco` is a English model originally trained by `hetpandya`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_small_tapaco_en_4.3.0_3.0_1675155980692.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_small_tapaco_en_4.3.0_3.0_1675155980692.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_small_tapaco","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_small_tapaco","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_small_tapaco| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|288.8 MB| + +## References + +- https://huggingface.co/hetpandya/t5-small-tapaco +- https://towardsdatascience.com/training-t5-for-paraphrase-generation-ab3b5be151a2 +- https://github.com/hetpandya +- https://www.linkedin.com/in/het-pandya \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_small_txtsql_en.md b/docs/_posts/Cabir40/2023-01-31-t5_small_txtsql_en.md new file mode 100644 index 00000000000000..c4ada5c24cd444 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_small_txtsql_en.md @@ -0,0 +1,86 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from NeuML) +author: John Snow Labs +name: t5_small_txtsql +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-small-txtsql` is a English model originally trained by `NeuML`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_small_txtsql_en_4.3.0_3.0_1675156028568.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_small_txtsql_en_4.3.0_3.0_1675156028568.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_small_txtsql","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_small_txtsql","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_small_txtsql| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|264.0 MB| + +## References + +- https://huggingface.co/NeuML/t5-small-txtsql +- https://github.com/neuml/txtai +- https://github.com/neuml/txtai/tree/master/models/txtsql \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_super_tiny_bahasa_cased_ms.md b/docs/_posts/Cabir40/2023-01-31-t5_super_tiny_bahasa_cased_ms.md new file mode 100644 index 00000000000000..fe7c7a0d86f8ed --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_super_tiny_bahasa_cased_ms.md @@ -0,0 +1,87 @@ +--- +layout: model +title: Malay T5ForConditionalGeneration Tiny Cased model (from mesolitica) +author: John Snow Labs +name: t5_super_tiny_bahasa_cased +date: 2023-01-31 +tags: [ms, open_source, t5, tensorflow] +task: Text Generation +language: ms +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-super-tiny-bahasa-cased` is a Malay model originally trained by `mesolitica`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_super_tiny_bahasa_cased_ms_4.3.0_3.0_1675156057502.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_super_tiny_bahasa_cased_ms_4.3.0_3.0_1675156057502.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_super_tiny_bahasa_cased","ms") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_super_tiny_bahasa_cased","ms") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_super_tiny_bahasa_cased| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|ms| +|Size:|40.5 MB| + +## References + +- https://huggingface.co/mesolitica/t5-super-tiny-bahasa-cased +- https://github.com/huseinzol05/malaya/tree/master/pretrained-model/t5/prepare +- https://github.com/google-research/text-to-text-transfer-transformer +- https://github.com/huseinzol05/Malaya/tree/master/pretrained-model/t5 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_tailor_en.md b/docs/_posts/Cabir40/2023-01-31-t5_tailor_en.md new file mode 100644 index 00000000000000..a5443ed2793aa5 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_tailor_en.md @@ -0,0 +1,86 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from allenai) +author: John Snow Labs +name: t5_tailor +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `tailor` is a English model originally trained by `allenai`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_tailor_en_4.3.0_3.0_1675157251689.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_tailor_en_4.3.0_3.0_1675157251689.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_tailor","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_tailor","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_tailor| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|926.0 MB| + +## References + +- https://huggingface.co/allenai/tailor +- https://homes.cs.washington.edu/~wtshuang/static/papers/2021-arxiv-tailor.pdf +- https://github.com/allenai/tailor \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_test_model_xx.md b/docs/_posts/Cabir40/2023-01-31-t5_test_model_xx.md new file mode 100644 index 00000000000000..08ca44a0f3faec --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_test_model_xx.md @@ -0,0 +1,85 @@ +--- +layout: model +title: Multilingual T5ForConditionalGeneration Cased model (from Lucapro) +author: John Snow Labs +name: t5_test_model +date: 2023-01-31 +tags: [en, ro, open_source, t5, xx, tensorflow] +task: Text Generation +language: xx +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `test-model` is a Multilingual model originally trained by `Lucapro`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_test_model_xx_4.3.0_3.0_1675157312429.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_test_model_xx_4.3.0_3.0_1675157312429.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_test_model","xx") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_test_model","xx") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_test_model| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|xx| +|Size:|260.2 MB| + +## References + +- https://huggingface.co/Lucapro/test-model +- https://paperswithcode.com/sota?task=Translation&dataset=wmt16+ro-en \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_text2sql_en.md b/docs/_posts/Cabir40/2023-01-31-t5_text2sql_en.md new file mode 100644 index 00000000000000..2527aa66d73514 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_text2sql_en.md @@ -0,0 +1,84 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from dsivakumar) +author: John Snow Labs +name: t5_text2sql +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `text2sql` is a English model originally trained by `dsivakumar`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_text2sql_en_4.3.0_3.0_1675157342237.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_text2sql_en_4.3.0_3.0_1675157342237.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_text2sql","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_text2sql","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_text2sql| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|283.3 MB| + +## References + +- https://huggingface.co/dsivakumar/text2sql \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_tiny_bahasa_cased_ms.md b/docs/_posts/Cabir40/2023-01-31-t5_tiny_bahasa_cased_ms.md new file mode 100644 index 00000000000000..8f16dcbdd5e2da --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_tiny_bahasa_cased_ms.md @@ -0,0 +1,87 @@ +--- +layout: model +title: Malay T5ForConditionalGeneration Tiny Cased model (from mesolitica) +author: John Snow Labs +name: t5_tiny_bahasa_cased +date: 2023-01-31 +tags: [ms, open_source, t5, tensorflow] +task: Text Generation +language: ms +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-tiny-bahasa-cased` is a Malay model originally trained by `mesolitica`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_tiny_bahasa_cased_ms_4.3.0_3.0_1675156097275.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_tiny_bahasa_cased_ms_4.3.0_3.0_1675156097275.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_tiny_bahasa_cased","ms") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_tiny_bahasa_cased","ms") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_tiny_bahasa_cased| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|ms| +|Size:|90.5 MB| + +## References + +- https://huggingface.co/mesolitica/t5-tiny-bahasa-cased +- https://github.com/huseinzol05/malaya/tree/master/pretrained-model/t5/prepare +- https://github.com/google-research/text-to-text-transfer-transformer +- https://github.com/huseinzol05/Malaya/tree/master/pretrained-model/t5 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_tiny_nl6_fi.md b/docs/_posts/Cabir40/2023-01-31-t5_tiny_nl6_fi.md new file mode 100644 index 00000000000000..0085d4f99fec6b --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_tiny_nl6_fi.md @@ -0,0 +1,100 @@ +--- +layout: model +title: Finnish T5ForConditionalGeneration Tiny Cased model (from Finnish-NLP) +author: John Snow Labs +name: t5_tiny_nl6 +date: 2023-01-31 +tags: [fi, open_source, t5, tensorflow] +task: Text Generation +language: fi +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-tiny-nl6-finnish` is a Finnish model originally trained by `Finnish-NLP`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_tiny_nl6_fi_4.3.0_3.0_1675156113232.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_tiny_nl6_fi_4.3.0_3.0_1675156113232.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_tiny_nl6","fi") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_tiny_nl6","fi") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_tiny_nl6| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|fi| +|Size:|145.8 MB| + +## References + +- https://huggingface.co/Finnish-NLP/t5-tiny-nl6-finnish +- https://arxiv.org/abs/1910.10683 +- https://github.com/google-research/text-to-text-transfer-transformer +- https://github.com/google-research/text-to-text-transfer-transformer/blob/main/released_checkpoints.md#t511 +- https://arxiv.org/abs/2002.05202 +- https://arxiv.org/abs/2109.10686 +- http://urn.fi/urn:nbn:fi:lb-2017070501 +- http://urn.fi/urn:nbn:fi:lb-2021050401 +- http://urn.fi/urn:nbn:fi:lb-2018121001 +- http://urn.fi/urn:nbn:fi:lb-2020021803 +- https://sites.research.google/trc/about/ +- https://github.com/google-research/t5x +- https://github.com/spyysalo/yle-corpus +- https://github.com/aajanki/eduskunta-vkk +- https://sites.research.google/trc/ +- https://www.linkedin.com/in/aapotanskanen/ +- https://www.linkedin.com/in/rasmustoivanen/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_titlewave_base_en.md b/docs/_posts/Cabir40/2023-01-31-t5_titlewave_base_en.md new file mode 100644 index 00000000000000..9b438ceade2995 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_titlewave_base_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from tennessejoyce) +author: John Snow Labs +name: t5_titlewave_base +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `titlewave-t5-base` is a English model originally trained by `tennessejoyce`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_titlewave_base_en_4.3.0_3.0_1675157398835.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_titlewave_base_en_4.3.0_3.0_1675157398835.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_titlewave_base","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_titlewave_base","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_titlewave_base| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|917.2 MB| + +## References + +- https://huggingface.co/tennessejoyce/titlewave-t5-base +- https://github.com/tennessejoyce/TitleWave +- https://github.com/tennessejoyce/TitleWave +- https://archive.org/details/stackexchange +- https://github.com/tennessejoyce/TitleWave/blob/master/model_training/test_summarizer.ipynb \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_translation_en2pt_xx.md b/docs/_posts/Cabir40/2023-01-31-t5_translation_en2pt_xx.md new file mode 100644 index 00000000000000..3d4e22e2e6e499 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_translation_en2pt_xx.md @@ -0,0 +1,86 @@ +--- +layout: model +title: Multilingual T5ForConditionalGeneration Cased model (from unicamp-dl) +author: John Snow Labs +name: t5_translation_en2pt +date: 2023-01-31 +tags: [pt, en, open_source, t5, xx, tensorflow] +task: Text Generation +language: xx +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `translation-en-pt-t5` is a Multilingual model originally trained by `unicamp-dl`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_translation_en2pt_xx_4.3.0_3.0_1675157481338.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_translation_en2pt_xx_4.3.0_3.0_1675157481338.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_translation_en2pt","xx") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_translation_en2pt","xx") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_translation_en2pt| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|xx| +|Size:|926.3 MB| + +## References + +- https://huggingface.co/unicamp-dl/translation-en-pt-t5 +- https://github.com/unicamp-dl/Lite-T5-Translation +- https://aclanthology.org/2020.wmt-1.90.pdf \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_translation_pt2en_xx.md b/docs/_posts/Cabir40/2023-01-31-t5_translation_pt2en_xx.md new file mode 100644 index 00000000000000..29bc57232d00ec --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_translation_pt2en_xx.md @@ -0,0 +1,86 @@ +--- +layout: model +title: Multilingual T5ForConditionalGeneration Cased model (from unicamp-dl) +author: John Snow Labs +name: t5_translation_pt2en +date: 2023-01-31 +tags: [pt, en, open_source, t5, xx, tensorflow] +task: Text Generation +language: xx +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `translation-pt-en-t5` is a Multilingual model originally trained by `unicamp-dl`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_translation_pt2en_xx_4.3.0_3.0_1675157565884.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_translation_pt2en_xx_4.3.0_3.0_1675157565884.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_translation_pt2en","xx") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_translation_pt2en","xx") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_translation_pt2en| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|xx| +|Size:|914.0 MB| + +## References + +- https://huggingface.co/unicamp-dl/translation-pt-en-t5 +- https://github.com/unicamp-dl/Lite-T5-Translation +- https://aclanthology.org/2020.wmt-1.90.pdf \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_triviaqa_base_en.md b/docs/_posts/Cabir40/2023-01-31-t5_triviaqa_base_en.md new file mode 100644 index 00000000000000..81e57e8404683a --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_triviaqa_base_en.md @@ -0,0 +1,87 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from deep-learning-analytics) +author: John Snow Labs +name: t5_triviaqa_base +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `triviaqa-t5-base` is a English model originally trained by `deep-learning-analytics`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_triviaqa_base_en_4.3.0_3.0_1675157650979.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_triviaqa_base_en_4.3.0_3.0_1675157650979.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_triviaqa_base","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_triviaqa_base","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_triviaqa_base| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|927.9 MB| + +## References + +- https://huggingface.co/deep-learning-analytics/triviaqa-t5-base +- https://medium.com/@priya.dwivedi/build-a-trivia-bot-using-t5-transformer-345ff83205b6 +- https://www.triviaquestionss.com/easy-trivia-questions/ +- https://laffgaff.com/easy-trivia-questions-and-answers/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_uk_summarizer_uk.md b/docs/_posts/Cabir40/2023-01-31-t5_uk_summarizer_uk.md new file mode 100644 index 00000000000000..516bcf0871f7ca --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_uk_summarizer_uk.md @@ -0,0 +1,84 @@ +--- +layout: model +title: Ukrainian T5ForConditionalGeneration Cased model (from ukr-models) +author: John Snow Labs +name: t5_uk_summarizer +date: 2023-01-31 +tags: [uk, open_source, t5, tensorflow] +task: Text Generation +language: uk +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `uk-summarizer` is a Ukrainian model originally trained by `ukr-models`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_uk_summarizer_uk_4.3.0_3.0_1675157739525.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_uk_summarizer_uk_4.3.0_3.0_1675157739525.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_uk_summarizer","uk") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_uk_summarizer","uk") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_uk_summarizer| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|uk| +|Size:|995.5 MB| + +## References + +- https://huggingface.co/ukr-models/uk-summarizer \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_unifiedqa_v2_base_1363200_en.md b/docs/_posts/Cabir40/2023-01-31-t5_unifiedqa_v2_base_1363200_en.md new file mode 100644 index 00000000000000..f17483ace9a98b --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_unifiedqa_v2_base_1363200_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from allenai) +author: John Snow Labs +name: t5_unifiedqa_v2_base_1363200 +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `unifiedqa-v2-t5-base-1363200` is a English model originally trained by `allenai`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_unifiedqa_v2_base_1363200_en_4.3.0_3.0_1675157943693.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_unifiedqa_v2_base_1363200_en_4.3.0_3.0_1675157943693.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_unifiedqa_v2_base_1363200","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_unifiedqa_v2_base_1363200","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_unifiedqa_v2_base_1363200| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|474.3 MB| + +## References + +- https://huggingface.co/allenai/unifiedqa-v2-t5-base-1363200 +- #further-details-httpsgithubcomallenaiunifiedqa +- https://github.com/allenai/unifiedqa +- #further-details-httpsgithubcomallenaiunifiedqa +- https://github.com/allenai/unifiedqa \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_unifiedqa_v2_small_1251000_en.md b/docs/_posts/Cabir40/2023-01-31-t5_unifiedqa_v2_small_1251000_en.md new file mode 100644 index 00000000000000..6fac1f3a91bc16 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_unifiedqa_v2_small_1251000_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from allenai) +author: John Snow Labs +name: t5_unifiedqa_v2_small_1251000 +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `unifiedqa-v2-t5-small-1251000` is a English model originally trained by `allenai`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_unifiedqa_v2_small_1251000_en_4.3.0_3.0_1675158019272.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_unifiedqa_v2_small_1251000_en_4.3.0_3.0_1675158019272.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_unifiedqa_v2_small_1251000","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_unifiedqa_v2_small_1251000","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_unifiedqa_v2_small_1251000| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|148.1 MB| + +## References + +- https://huggingface.co/allenai/unifiedqa-v2-t5-small-1251000 +- #further-details-httpsgithubcomallenaiunifiedqa +- https://github.com/allenai/unifiedqa +- #further-details-httpsgithubcomallenaiunifiedqa +- https://github.com/allenai/unifiedqa \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_unifiedqa_v2_small_1363200_en.md b/docs/_posts/Cabir40/2023-01-31-t5_unifiedqa_v2_small_1363200_en.md new file mode 100644 index 00000000000000..0da85fa507823e --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_unifiedqa_v2_small_1363200_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from allenai) +author: John Snow Labs +name: t5_unifiedqa_v2_small_1363200 +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `unifiedqa-v2-t5-small-1363200` is a English model originally trained by `allenai`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_unifiedqa_v2_small_1363200_en_4.3.0_3.0_1675158081810.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_unifiedqa_v2_small_1363200_en_4.3.0_3.0_1675158081810.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_unifiedqa_v2_small_1363200","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_unifiedqa_v2_small_1363200","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_unifiedqa_v2_small_1363200| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|148.1 MB| + +## References + +- https://huggingface.co/allenai/unifiedqa-v2-t5-small-1363200 +- #further-details-httpsgithubcomallenaiunifiedqa +- https://github.com/allenai/unifiedqa +- #further-details-httpsgithubcomallenaiunifiedqa +- https://github.com/allenai/unifiedqa \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_v1_1_base_en.md b/docs/_posts/Cabir40/2023-01-31-t5_v1_1_base_en.md new file mode 100644 index 00000000000000..724328fc614355 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_v1_1_base_en.md @@ -0,0 +1,89 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from google) +author: John Snow Labs +name: t5_v1_1_base +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-v1_1-base` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_v1_1_base_en_4.3.0_3.0_1675156307761.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_v1_1_base_en_4.3.0_3.0_1675156307761.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_v1_1_base","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_v1_1_base","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_v1_1_base| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|521.2 MB| + +## References + +- https://huggingface.co/google/t5-v1_1-base +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://github.com/google-research/text-to-text-transfer-transformer/blob/master/released_checkpoints.md#t511 +- https://arxiv.org/abs/2002.05202 +- https://arxiv.org/pdf/1910.10683.pdf +- https://camo.githubusercontent.com/623b4dea0b653f2ad3f36c71ebfe749a677ac0a1/68747470733a2f2f6d69726f2e6d656469756d2e636f6d2f6d61782f343030362f312a44304a31674e51663876727255704b657944387750412e706e67 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_v1_1_small_en.md b/docs/_posts/Cabir40/2023-01-31-t5_v1_1_small_en.md new file mode 100644 index 00000000000000..2c90b2da8c837d --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_v1_1_small_en.md @@ -0,0 +1,89 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from google) +author: John Snow Labs +name: t5_v1_1_small +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-v1_1-small` is a English model originally trained by `google`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_v1_1_small_en_4.3.0_3.0_1675156397272.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_v1_1_small_en_4.3.0_3.0_1675156397272.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_v1_1_small","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_v1_1_small","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_v1_1_small| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|179.4 MB| + +## References + +- https://huggingface.co/google/t5-v1_1-small +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://github.com/google-research/text-to-text-transfer-transformer/blob/master/released_checkpoints.md#t511 +- https://arxiv.org/abs/2002.05202 +- https://arxiv.org/pdf/1910.10683.pdf +- https://camo.githubusercontent.com/623b4dea0b653f2ad3f36c71ebfe749a677ac0a1/68747470733a2f2f6d69726f2e6d656469756d2e636f6d2f6d61782f343030362f312a44304a31674e51663876727255704b657944387750412e706e67 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_vi_small_vi.md b/docs/_posts/Cabir40/2023-01-31-t5_vi_small_vi.md new file mode 100644 index 00000000000000..423b8d1cb4524f --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_vi_small_vi.md @@ -0,0 +1,84 @@ +--- +layout: model +title: Vietnamese T5ForConditionalGeneration Small Cased model (from NlpHUST) +author: John Snow Labs +name: t5_vi_small +date: 2023-01-31 +tags: [vi, open_source, t5, tensorflow] +task: Text Generation +language: vi +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-vi-en-small` is a Vietnamese model originally trained by `NlpHUST`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_vi_small_vi_4.3.0_3.0_1675156710542.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_vi_small_vi_4.3.0_3.0_1675156710542.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_vi_small","vi") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_vi_small","vi") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_vi_small| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|vi| +|Size:|819.8 MB| + +## References + +- https://huggingface.co/NlpHUST/t5-vi-en-small \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_vit5_base_vi.md b/docs/_posts/Cabir40/2023-01-31-t5_vit5_base_vi.md new file mode 100644 index 00000000000000..a3adf0425160fb --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_vit5_base_vi.md @@ -0,0 +1,85 @@ +--- +layout: model +title: Vietnamese T5ForConditionalGeneration Base Cased model (from VietAI) +author: John Snow Labs +name: t5_vit5_base +date: 2023-01-31 +tags: [vi, open_source, t5, tensorflow] +task: Text Generation +language: vi +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `vit5-base` is a Vietnamese model originally trained by `VietAI`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_vit5_base_vi_4.3.0_3.0_1675158262308.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_vit5_base_vi_4.3.0_3.0_1675158262308.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_vit5_base","vi") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_vit5_base","vi") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_vit5_base| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|vi| +|Size:|485.0 MB| + +## References + +- https://huggingface.co/VietAI/vit5-base +- https://github.com/vietai/ViT5 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_vit5_base_vietnews_summarization_vi.md b/docs/_posts/Cabir40/2023-01-31-t5_vit5_base_vietnews_summarization_vi.md new file mode 100644 index 00000000000000..fce96de70c8d8f --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_vit5_base_vietnews_summarization_vi.md @@ -0,0 +1,87 @@ +--- +layout: model +title: Vietnamese T5ForConditionalGeneration Base Cased model (from VietAI) +author: John Snow Labs +name: t5_vit5_base_vietnews_summarization +date: 2023-01-31 +tags: [vi, open_source, t5, tensorflow] +task: Text Generation +language: vi +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `vit5-base-vietnews-summarization` is a Vietnamese model originally trained by `VietAI`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_vit5_base_vietnews_summarization_vi_4.3.0_3.0_1675158454304.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_vit5_base_vietnews_summarization_vi_4.3.0_3.0_1675158454304.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_vit5_base_vietnews_summarization","vi") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_vit5_base_vietnews_summarization","vi") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_vit5_base_vietnews_summarization| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|vi| +|Size:|485.0 MB| + +## References + +- https://huggingface.co/VietAI/vit5-base-vietnews-summarization +- https://paperswithcode.com/sota/abstractive-text-summarization-on-vietnews?p=vit5-pretrained-text-to-text-transformer-for +- https://github.com/vietai/ViT5 +- https://github.com/vietai/ViT5/blob/main/eval/Eval_vietnews_sum.ipynb \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_vlt5_base_keywords_xx.md b/docs/_posts/Cabir40/2023-01-31-t5_vlt5_base_keywords_xx.md new file mode 100644 index 00000000000000..726a9081b60f27 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_vlt5_base_keywords_xx.md @@ -0,0 +1,88 @@ +--- +layout: model +title: Multilingual T5ForConditionalGeneration Base Cased model (from Voicelab) +author: John Snow Labs +name: t5_vlt5_base_keywords +date: 2023-01-31 +tags: [en, pl, open_source, t5, xx, tensorflow] +task: Text Generation +language: xx +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `vlt5-base-keywords` is a Multilingual model originally trained by `Voicelab`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_vlt5_base_keywords_xx_4.3.0_3.0_1675158538277.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_vlt5_base_keywords_xx_4.3.0_3.0_1675158538277.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_vlt5_base_keywords","xx") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_vlt5_base_keywords","xx") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_vlt5_base_keywords| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|xx| +|Size:|1.1 GB| + +## References + +- https://huggingface.co/Voicelab/vlt5-base-keywords +- https://nlp-demo-1.voicelab.ai/ +- https://arxiv.org/abs/2209.14008 +- https://arxiv.org/abs/2209.14008 +- https://voicelab.ai/contact/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_weighter_cnndm_en.md b/docs/_posts/Cabir40/2023-01-31-t5_weighter_cnndm_en.md new file mode 100644 index 00000000000000..d9ceeedb7a87a7 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_weighter_cnndm_en.md @@ -0,0 +1,86 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from ThomasNLG) +author: John Snow Labs +name: t5_weighter_cnndm +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5-weighter_cnndm-en` is a English model originally trained by `ThomasNLG`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_weighter_cnndm_en_4.3.0_3.0_1675156764080.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_weighter_cnndm_en_4.3.0_3.0_1675156764080.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_weighter_cnndm","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_weighter_cnndm","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_weighter_cnndm| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|277.8 MB| + +## References + +- https://huggingface.co/ThomasNLG/t5-weighter_cnndm-en +- https://github.com/ThomasScialom/QuestEval +- https://arxiv.org/abs/2103.12693 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_wikihow_small_en.md b/docs/_posts/Cabir40/2023-01-31-t5_wikihow_small_en.md new file mode 100644 index 00000000000000..9b0ba9ac116fc7 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_wikihow_small_en.md @@ -0,0 +1,85 @@ +--- +layout: model +title: English T5ForConditionalGeneration Small Cased model (from deep-learning-analytics) +author: John Snow Labs +name: t5_wikihow_small +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `wikihow-t5-small` is a English model originally trained by `deep-learning-analytics`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_wikihow_small_en_4.3.0_3.0_1675158602814.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_wikihow_small_en_4.3.0_3.0_1675158602814.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_wikihow_small","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_wikihow_small","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_wikihow_small| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|286.9 MB| + +## References + +- https://huggingface.co/deep-learning-analytics/wikihow-t5-small +- https://medium.com/@priya.dwivedi/fine-tuning-a-t5-transformer-for-any-summarization-task-82334c64c81 \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_wikisql_en2sql_en.md b/docs/_posts/Cabir40/2023-01-31-t5_wikisql_en2sql_en.md new file mode 100644 index 00000000000000..099ae1b2322e3e --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_wikisql_en2sql_en.md @@ -0,0 +1,87 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from dbernsohn) +author: John Snow Labs +name: t5_wikisql_en2sql +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5_wikisql_en2SQL` is a English model originally trained by `dbernsohn`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_wikisql_en2sql_en_4.3.0_3.0_1675157192158.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_wikisql_en2sql_en_4.3.0_3.0_1675157192158.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_wikisql_en2sql","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_wikisql_en2sql","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_wikisql_en2sql| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|288.2 MB| + +## References + +- https://huggingface.co/dbernsohn/t5_wikisql_en2SQL +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://github.com/DorBernsohn/CodeLM/tree/main/SQLM +- https://www.linkedin.com/in/dor-bernsohn-70b2b1146/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_wikisql_sql2en_en.md b/docs/_posts/Cabir40/2023-01-31-t5_wikisql_sql2en_en.md new file mode 100644 index 00000000000000..2c1f448d4f65a3 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_wikisql_sql2en_en.md @@ -0,0 +1,87 @@ +--- +layout: model +title: English T5ForConditionalGeneration Cased model (from dbernsohn) +author: John Snow Labs +name: t5_wikisql_sql2en +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `t5_wikisql_SQL2en` is a English model originally trained by `dbernsohn`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_wikisql_sql2en_en_4.3.0_3.0_1675157160888.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_wikisql_sql2en_en_4.3.0_3.0_1675157160888.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_wikisql_sql2en","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_wikisql_sql2en","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_wikisql_sql2en| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|288.2 MB| + +## References + +- https://huggingface.co/dbernsohn/t5_wikisql_SQL2en +- https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html +- https://github.com/DorBernsohn/CodeLM/tree/main/SQLM +- https://www.linkedin.com/in/dor-bernsohn-70b2b1146/ \ No newline at end of file diff --git a/docs/_posts/Cabir40/2023-01-31-t5_yahoo_answers_base_v1_en.md b/docs/_posts/Cabir40/2023-01-31-t5_yahoo_answers_base_v1_en.md new file mode 100644 index 00000000000000..3635d5cba01e03 --- /dev/null +++ b/docs/_posts/Cabir40/2023-01-31-t5_yahoo_answers_base_v1_en.md @@ -0,0 +1,89 @@ +--- +layout: model +title: English T5ForConditionalGeneration Base Cased model (from doc2query) +author: John Snow Labs +name: t5_yahoo_answers_base_v1 +date: 2023-01-31 +tags: [en, open_source, t5, tensorflow] +task: Text Generation +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: T5Transformer +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained T5ForConditionalGeneration model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `yahoo_answers-t5-base-v1` is a English model originally trained by `doc2query`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/t5_yahoo_answers_base_v1_en_4.3.0_3.0_1675158667385.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/t5_yahoo_answers_base_v1_en_4.3.0_3.0_1675158667385.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCols("text") \ + .setOutputCols("document") + +t5 = T5Transformer.pretrained("t5_yahoo_answers_base_v1","en") \ + .setInputCols("document") \ + .setOutputCol("answers") + +pipeline = Pipeline(stages=[documentAssembler, t5]) + +data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` +```scala +val documentAssembler = new DocumentAssembler() + .setInputCols("text") + .setOutputCols("document") + +val t5 = T5Transformer.pretrained("t5_yahoo_answers_base_v1","en") + .setInputCols("document") + .setOutputCol("answers") + +val pipeline = new Pipeline().setStages(Array(documentAssembler, t5)) + +val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|t5_yahoo_answers_base_v1| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[documents]| +|Output Labels:|[t5]| +|Language:|en| +|Size:|1.0 GB| + +## References + +- https://huggingface.co/doc2query/yahoo_answers-t5-base-v1 +- https://arxiv.org/abs/1904.08375 +- https://cs.uwaterloo.ca/~jimmylin/publications/Nogueira_Lin_2019_docTTTTTquery-v2.pdf +- https://arxiv.org/abs/2104.08663 +- https://github.com/UKPLab/beir +- https://www.sbert.net/examples/unsupervised_learning/query_generation/README.html \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_01_dialdoc_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_01_dialdoc_en.md new file mode 100644 index 00000000000000..4f5dd7e272f856 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_01_dialdoc_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from alistvt) +author: John Snow Labs +name: roberta_qa_01_dialdoc +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `01-roberta-dialdoc` is a English model originally trained by `alistvt`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_01_dialdoc_en_4.3.0_3.0_1674206907196.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_01_dialdoc_en_4.3.0_3.0_1674206907196.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_01_dialdoc","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_01_dialdoc","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_01_dialdoc| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|1.3 GB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/alistvt/01-roberta-dialdoc \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_addi_fr_xlm_r_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_addi_fr_xlm_r_en.md new file mode 100644 index 00000000000000..9ed066eea6af33 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_addi_fr_xlm_r_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from Gantenbein) +author: John Snow Labs +name: roberta_qa_addi_fr_xlm_r +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `ADDI-FR-XLM-R` is a English model originally trained by `Gantenbein`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_addi_fr_xlm_r_en_4.3.0_3.0_1674207724209.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_addi_fr_xlm_r_en_4.3.0_3.0_1674207724209.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_addi_fr_xlm_r","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_addi_fr_xlm_r","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_addi_fr_xlm_r| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|422.7 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Gantenbein/ADDI-FR-XLM-R \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_ai_club_inductions_21_nlp_base_squad_v2_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_ai_club_inductions_21_nlp_base_squad_v2_en.md new file mode 100644 index 00000000000000..0972b0a7b3c365 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_ai_club_inductions_21_nlp_base_squad_v2_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from AyushPJ) +author: John Snow Labs +name: roberta_qa_ai_club_inductions_21_nlp_base_squad_v2 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `ai-club-inductions-21-nlp-roBERTa-base-squad-v2` is a English model originally trained by `AyushPJ`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_ai_club_inductions_21_nlp_base_squad_v2_en_4.3.0_3.0_1674209021708.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_ai_club_inductions_21_nlp_base_squad_v2_en_4.3.0_3.0_1674209021708.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_ai_club_inductions_21_nlp_base_squad_v2","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_ai_club_inductions_21_nlp_base_squad_v2","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_ai_club_inductions_21_nlp_base_squad_v2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|465.3 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AyushPJ/ai-club-inductions-21-nlp-roBERTa-base-squad-v2 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_ai_club_inductions_21_nlp_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_ai_club_inductions_21_nlp_en.md new file mode 100644 index 00000000000000..79b8c1b716df96 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_ai_club_inductions_21_nlp_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from AyushPJ) +author: John Snow Labs +name: roberta_qa_ai_club_inductions_21_nlp +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `ai-club-inductions-21-nlp-roBERTa` is a English model originally trained by `AyushPJ`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_ai_club_inductions_21_nlp_en_4.3.0_3.0_1674208962596.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_ai_club_inductions_21_nlp_en_4.3.0_3.0_1674208962596.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_ai_club_inductions_21_nlp","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_ai_club_inductions_21_nlp","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_ai_club_inductions_21_nlp| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|465.3 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AyushPJ/ai-club-inductions-21-nlp-roBERTa \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_amitjohn007_base_finetuned_squad_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_amitjohn007_base_finetuned_squad_en.md new file mode 100644 index 00000000000000..f9f1d5393bc23d --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_amitjohn007_base_finetuned_squad_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from amitjohn007) +author: John Snow Labs +name: roberta_qa_amitjohn007_base_finetuned_squad +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-finetuned-squad` is a English model originally trained by `amitjohn007`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_amitjohn007_base_finetuned_squad_en_4.3.0_3.0_1674217120272.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_amitjohn007_base_finetuned_squad_en_4.3.0_3.0_1674217120272.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_amitjohn007_base_finetuned_squad","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_amitjohn007_base_finetuned_squad","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_amitjohn007_base_finetuned_squad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.3 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/amitjohn007/roberta-base-finetuned-squad \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_autotrain_test2_756523213_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_autotrain_test2_756523213_en.md new file mode 100644 index 00000000000000..9029ea983e5348 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_autotrain_test2_756523213_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from AlirezaBaneshi) +author: John Snow Labs +name: roberta_qa_autotrain_test2_756523213 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `autotrain-test2-756523213` is a English model originally trained by `AlirezaBaneshi`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_autotrain_test2_756523213_en_4.3.0_3.0_1674209108948.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_autotrain_test2_756523213_en_4.3.0_3.0_1674209108948.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_autotrain_test2_756523213","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_autotrain_test2_756523213","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_autotrain_test2_756523213| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|415.9 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AlirezaBaneshi/autotrain-test2-756523213 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_autotrain_test2_756523214_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_autotrain_test2_756523214_en.md new file mode 100644 index 00000000000000..8d3eb536e920c8 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_autotrain_test2_756523214_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from AlirezaBaneshi) +author: John Snow Labs +name: roberta_qa_autotrain_test2_756523214 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `autotrain-test2-756523214` is a English model originally trained by `AlirezaBaneshi`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_autotrain_test2_756523214_en_4.3.0_3.0_1674209197246.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_autotrain_test2_756523214_en_4.3.0_3.0_1674209197246.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_autotrain_test2_756523214","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_autotrain_test2_756523214","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_autotrain_test2_756523214| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|415.9 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AlirezaBaneshi/autotrain-test2-756523214 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_avioo1_base_squad2_finetuned_squad_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_avioo1_base_squad2_finetuned_squad_en.md new file mode 100644 index 00000000000000..a817ac110154d3 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_avioo1_base_squad2_finetuned_squad_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from avioo1) +author: John Snow Labs +name: roberta_qa_avioo1_base_squad2_finetuned_squad +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-squad2-finetuned-squad` is a English model originally trained by `avioo1`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_avioo1_base_squad2_finetuned_squad_en_4.3.0_3.0_1674219191405.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_avioo1_base_squad2_finetuned_squad_en_4.3.0_3.0_1674219191405.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_avioo1_base_squad2_finetuned_squad","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_avioo1_base_squad2_finetuned_squad","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_avioo1_base_squad2_finetuned_squad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.4 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/avioo1/roberta-base-squad2-finetuned-squad \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_attribute_correction_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_attribute_correction_en.md new file mode 100644 index 00000000000000..f6330a2ec2a421 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_attribute_correction_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from ksabeh) +author: John Snow Labs +name: roberta_qa_base_attribute_correction +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-attribute-correction` is a English model originally trained by `ksabeh`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_attribute_correction_en_4.3.0_3.0_1674212650491.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_attribute_correction_en_4.3.0_3.0_1674212650491.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_attribute_correction","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_attribute_correction","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_attribute_correction| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|430.8 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/ksabeh/roberta-base-attribute-correction \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_attribute_correction_mlm_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_attribute_correction_mlm_en.md new file mode 100644 index 00000000000000..1e4cb337f7d72c --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_attribute_correction_mlm_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from ksabeh) +author: John Snow Labs +name: roberta_qa_base_attribute_correction_mlm +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-attribute-correction-mlm` is a English model originally trained by `ksabeh`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_attribute_correction_mlm_en_4.3.0_3.0_1674212707495.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_attribute_correction_mlm_en_4.3.0_3.0_1674212707495.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_attribute_correction_mlm","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_attribute_correction_mlm","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_attribute_correction_mlm| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|466.8 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/ksabeh/roberta-base-attribute-correction-mlm \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_attribute_correction_mlm_titles_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_attribute_correction_mlm_titles_en.md new file mode 100644 index 00000000000000..33c1d239a47819 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_attribute_correction_mlm_titles_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from ksabeh) +author: John Snow Labs +name: roberta_qa_base_attribute_correction_mlm_titles +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-attribute-correction-mlm-titles` is a English model originally trained by `ksabeh`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_attribute_correction_mlm_titles_en_4.3.0_3.0_1674212765867.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_attribute_correction_mlm_titles_en_4.3.0_3.0_1674212765867.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_attribute_correction_mlm_titles","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_attribute_correction_mlm_titles","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_attribute_correction_mlm_titles| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|466.8 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/ksabeh/roberta-base-attribute-correction-mlm-titles \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_best_squad2_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_best_squad2_en.md new file mode 100644 index 00000000000000..9aed0b4d745581 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_best_squad2_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from PremalMatalia) +author: John Snow Labs +name: roberta_qa_base_best_squad2 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-best-squad2` is a English model originally trained by `PremalMatalia`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_best_squad2_en_4.3.0_3.0_1674212826277.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_best_squad2_en_4.3.0_3.0_1674212826277.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_best_squad2","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_best_squad2","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_best_squad2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.3 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/PremalMatalia/roberta-base-best-squad2 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_bne_becas_es.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_bne_becas_es.md new file mode 100644 index 00000000000000..646cae129be7af --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_bne_becas_es.md @@ -0,0 +1,88 @@ +--- +layout: model +title: Spanish RobertaForQuestionAnswering Base Cased model (from Evelyn18) +author: John Snow Labs +name: roberta_qa_base_bne_becas +date: 2023-01-20 +tags: [es, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: es +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-bne-ROBERTaBECAS` is a Spanish model originally trained by `Evelyn18`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_bne_becas_es_4.3.0_3.0_1674212894934.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_bne_becas_es_4.3.0_3.0_1674212894934.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_bne_becas","es")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_bne_becas","es") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_bne_becas| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|es| +|Size:|420.6 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Evelyn18/roberta-base-bne-ROBERTaBECAS \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_ca_v2_catalan_ca.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_ca_v2_catalan_ca.md new file mode 100644 index 00000000000000..b9f8b4271307a9 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_ca_v2_catalan_ca.md @@ -0,0 +1,88 @@ +--- +layout: model +title: Catalan RobertaForQuestionAnswering Base Cased model (from crodri) +author: John Snow Labs +name: roberta_qa_base_ca_v2_catalan +date: 2023-01-20 +tags: [ca, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: ca +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-ca-v2-qa-catalanqa` is a Catalan model originally trained by `crodri`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_ca_v2_catalan_ca_4.3.0_3.0_1674213071415.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_ca_v2_catalan_ca_4.3.0_3.0_1674213071415.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_ca_v2_catalan","ca")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_ca_v2_catalan","ca") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_ca_v2_catalan| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|ca| +|Size:|456.5 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/crodri/roberta-base-ca-v2-qa-catalanqa \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_chaii_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_chaii_en.md new file mode 100644 index 00000000000000..a40edd128b8122 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_chaii_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from SauravMaheshkar) +author: John Snow Labs +name: roberta_qa_base_chaii +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-chaii` is a English model originally trained by `SauravMaheshkar`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_chaii_en_4.3.0_3.0_1674213133232.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_chaii_en_4.3.0_3.0_1674213133232.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_chaii","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_chaii","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_chaii| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.7 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/SauravMaheshkar/roberta-base-chaii \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_custom_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_custom_en.md new file mode 100644 index 00000000000000..25f8235413595b --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_custom_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from eAsyle) +author: John Snow Labs +name: roberta_qa_base_custom +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta_base_custom_QA` is a English model originally trained by `eAsyle`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_custom_en_4.3.0_3.0_1674223065625.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_custom_en_4.3.0_3.0_1674223065625.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_custom","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_custom","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_custom| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|424.1 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/eAsyle/roberta_base_custom_QA \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_emr_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_emr_en.md new file mode 100644 index 00000000000000..dffd6f0304e0a6 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_emr_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from emr-se-miniproject) +author: John Snow Labs +name: roberta_qa_base_emr +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-emr` is a English model originally trained by `emr-se-miniproject`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_emr_en_4.3.0_3.0_1674213227451.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_emr_en_4.3.0_3.0_1674213227451.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_emr","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_emr","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_emr| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.5 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/emr-se-miniproject/roberta-base-emr \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_0_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_0_en.md new file mode 100644 index 00000000000000..9f017ef292fa19 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_0_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_0 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-1024-finetuned-squad-seed-0` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_0_en_4.3.0_3.0_1674213299788.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_0_en_4.3.0_3.0_1674213299788.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_0","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_0","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_0| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|439.5 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-1024-finetuned-squad-seed-0 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_10_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_10_en.md new file mode 100644 index 00000000000000..7fe6671a02cce0 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_10_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_10 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-1024-finetuned-squad-seed-10` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_10_en_4.3.0_3.0_1674213377790.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_10_en_4.3.0_3.0_1674213377790.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_10","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_10","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_10| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|439.4 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-1024-finetuned-squad-seed-10 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_42_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_42_en.md new file mode 100644 index 00000000000000..fbf4d9eda1164e --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_42_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_42 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-1024-finetuned-squad-seed-42` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_42_en_4.3.0_3.0_1674213511887.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_42_en_4.3.0_3.0_1674213511887.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_42","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_42","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_42| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|447.1 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-1024-finetuned-squad-seed-42 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_4_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_4_en.md new file mode 100644 index 00000000000000..f1aecdc7cf42e2 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_4_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_4 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-1024-finetuned-squad-seed-4` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_4_en_4.3.0_3.0_1674213450529.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_4_en_4.3.0_3.0_1674213450529.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_4","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_4","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_4| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|439.5 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-1024-finetuned-squad-seed-4 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_6_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_6_en.md new file mode 100644 index 00000000000000..888fc29b4049ea --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_6_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_6 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-1024-finetuned-squad-seed-6` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_6_en_4.3.0_3.0_1674213584769.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_6_en_4.3.0_3.0_1674213584769.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_6","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_6","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_1024_finetuned_squad_seed_6| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|439.4 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-1024-finetuned-squad-seed-6 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_128_finetuned_squad_seed_10_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_128_finetuned_squad_seed_10_en.md new file mode 100644 index 00000000000000..be3987e0c11291 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_128_finetuned_squad_seed_10_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_128_finetuned_squad_seed_10 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-128-finetuned-squad-seed-10` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_128_finetuned_squad_seed_10_en_4.3.0_3.0_1674213671532.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_128_finetuned_squad_seed_10_en_4.3.0_3.0_1674213671532.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_128_finetuned_squad_seed_10","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_128_finetuned_squad_seed_10","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_128_finetuned_squad_seed_10| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|423.1 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-128-finetuned-squad-seed-10 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_128_finetuned_squad_seed_2_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_128_finetuned_squad_seed_2_en.md new file mode 100644 index 00000000000000..c7b0fc6de86781 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_128_finetuned_squad_seed_2_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_128_finetuned_squad_seed_2 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-128-finetuned-squad-seed-2` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_128_finetuned_squad_seed_2_en_4.3.0_3.0_1674213755024.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_128_finetuned_squad_seed_2_en_4.3.0_3.0_1674213755024.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_128_finetuned_squad_seed_2","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_128_finetuned_squad_seed_2","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_128_finetuned_squad_seed_2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|423.1 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-128-finetuned-squad-seed-2 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_128_finetuned_squad_seed_42_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_128_finetuned_squad_seed_42_en.md new file mode 100644 index 00000000000000..c5b01ea1ea39bc --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_128_finetuned_squad_seed_42_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_128_finetuned_squad_seed_42 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-128-finetuned-squad-seed-42` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_128_finetuned_squad_seed_42_en_4.3.0_3.0_1674213900898.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_128_finetuned_squad_seed_42_en_4.3.0_3.0_1674213900898.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_128_finetuned_squad_seed_42","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_128_finetuned_squad_seed_42","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_128_finetuned_squad_seed_42| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|431.0 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-128-finetuned-squad-seed-42 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_128_finetuned_squad_seed_4_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_128_finetuned_squad_seed_4_en.md new file mode 100644 index 00000000000000..45b1313526488f --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_128_finetuned_squad_seed_4_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_128_finetuned_squad_seed_4 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-128-finetuned-squad-seed-4` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_128_finetuned_squad_seed_4_en_4.3.0_3.0_1674213839010.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_128_finetuned_squad_seed_4_en_4.3.0_3.0_1674213839010.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_128_finetuned_squad_seed_4","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_128_finetuned_squad_seed_4","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_128_finetuned_squad_seed_4| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|423.2 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-128-finetuned-squad-seed-4 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_128_finetuned_squad_seed_6_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_128_finetuned_squad_seed_6_en.md new file mode 100644 index 00000000000000..c01c6f8871b98f --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_128_finetuned_squad_seed_6_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_128_finetuned_squad_seed_6 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-128-finetuned-squad-seed-6` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_128_finetuned_squad_seed_6_en_4.3.0_3.0_1674213983376.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_128_finetuned_squad_seed_6_en_4.3.0_3.0_1674213983376.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_128_finetuned_squad_seed_6","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_128_finetuned_squad_seed_6","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_128_finetuned_squad_seed_6| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|422.7 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-128-finetuned-squad-seed-6 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_128_finetuned_squad_seed_8_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_128_finetuned_squad_seed_8_en.md new file mode 100644 index 00000000000000..4ff5553543b735 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_128_finetuned_squad_seed_8_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_128_finetuned_squad_seed_8 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-128-finetuned-squad-seed-8` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_128_finetuned_squad_seed_8_en_4.3.0_3.0_1674214068825.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_128_finetuned_squad_seed_8_en_4.3.0_3.0_1674214068825.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_128_finetuned_squad_seed_8","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_128_finetuned_squad_seed_8","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_128_finetuned_squad_seed_8| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|423.0 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-128-finetuned-squad-seed-8 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_16_finetuned_squad_seed_0_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_16_finetuned_squad_seed_0_en.md new file mode 100644 index 00000000000000..48587f12544e1c --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_16_finetuned_squad_seed_0_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_16_finetuned_squad_seed_0 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-16-finetuned-squad-seed-0` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_16_finetuned_squad_seed_0_en_4.3.0_3.0_1674214155549.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_16_finetuned_squad_seed_0_en_4.3.0_3.0_1674214155549.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_16_finetuned_squad_seed_0","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_16_finetuned_squad_seed_0","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_16_finetuned_squad_seed_0| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|416.0 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-16-finetuned-squad-seed-0 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_16_finetuned_squad_seed_10_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_16_finetuned_squad_seed_10_en.md new file mode 100644 index 00000000000000..de3f1ad937c5b8 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_16_finetuned_squad_seed_10_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_16_finetuned_squad_seed_10 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-16-finetuned-squad-seed-10` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_16_finetuned_squad_seed_10_en_4.3.0_3.0_1674214242487.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_16_finetuned_squad_seed_10_en_4.3.0_3.0_1674214242487.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_16_finetuned_squad_seed_10","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_16_finetuned_squad_seed_10","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_16_finetuned_squad_seed_10| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|416.2 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-16-finetuned-squad-seed-10 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_16_finetuned_squad_seed_2_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_16_finetuned_squad_seed_2_en.md new file mode 100644 index 00000000000000..0c8173ae18249f --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_16_finetuned_squad_seed_2_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_16_finetuned_squad_seed_2 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-16-finetuned-squad-seed-2` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_16_finetuned_squad_seed_2_en_4.3.0_3.0_1674214328836.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_16_finetuned_squad_seed_2_en_4.3.0_3.0_1674214328836.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_16_finetuned_squad_seed_2","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_16_finetuned_squad_seed_2","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_16_finetuned_squad_seed_2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|416.2 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-16-finetuned-squad-seed-2 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_16_finetuned_squad_seed_42_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_16_finetuned_squad_seed_42_en.md new file mode 100644 index 00000000000000..18304b62464fed --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_16_finetuned_squad_seed_42_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_16_finetuned_squad_seed_42 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-16-finetuned-squad-seed-42` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_16_finetuned_squad_seed_42_en_4.3.0_3.0_1674214395082.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_16_finetuned_squad_seed_42_en_4.3.0_3.0_1674214395082.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_16_finetuned_squad_seed_42","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_16_finetuned_squad_seed_42","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_16_finetuned_squad_seed_42| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|425.0 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-16-finetuned-squad-seed-42 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_16_finetuned_squad_seed_6_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_16_finetuned_squad_seed_6_en.md new file mode 100644 index 00000000000000..6783c3a8106fb3 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_16_finetuned_squad_seed_6_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_16_finetuned_squad_seed_6 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-16-finetuned-squad-seed-6` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_16_finetuned_squad_seed_6_en_4.3.0_3.0_1674214482569.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_16_finetuned_squad_seed_6_en_4.3.0_3.0_1674214482569.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_16_finetuned_squad_seed_6","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_16_finetuned_squad_seed_6","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_16_finetuned_squad_seed_6| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|416.1 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-16-finetuned-squad-seed-6 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_16_finetuned_squad_seed_8_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_16_finetuned_squad_seed_8_en.md new file mode 100644 index 00000000000000..a554c02bb62dd1 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_16_finetuned_squad_seed_8_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_16_finetuned_squad_seed_8 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-16-finetuned-squad-seed-8` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_16_finetuned_squad_seed_8_en_4.3.0_3.0_1674214569437.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_16_finetuned_squad_seed_8_en_4.3.0_3.0_1674214569437.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_16_finetuned_squad_seed_8","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_16_finetuned_squad_seed_8","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_16_finetuned_squad_seed_8| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|415.9 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-16-finetuned-squad-seed-8 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_256_finetuned_squad_seed_0_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_256_finetuned_squad_seed_0_en.md new file mode 100644 index 00000000000000..3e8cf20de9718e --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_256_finetuned_squad_seed_0_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_256_finetuned_squad_seed_0 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-256-finetuned-squad-seed-0` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_256_finetuned_squad_seed_0_en_4.3.0_3.0_1674214651872.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_256_finetuned_squad_seed_0_en_4.3.0_3.0_1674214651872.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_256_finetuned_squad_seed_0","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_256_finetuned_squad_seed_0","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_256_finetuned_squad_seed_0| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|427.2 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-256-finetuned-squad-seed-0 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_256_finetuned_squad_seed_10_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_256_finetuned_squad_seed_10_en.md new file mode 100644 index 00000000000000..e4f08e1de6b865 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_256_finetuned_squad_seed_10_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_256_finetuned_squad_seed_10 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-256-finetuned-squad-seed-10` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_256_finetuned_squad_seed_10_en_4.3.0_3.0_1674214731317.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_256_finetuned_squad_seed_10_en_4.3.0_3.0_1674214731317.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_256_finetuned_squad_seed_10","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_256_finetuned_squad_seed_10","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_256_finetuned_squad_seed_10| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|427.3 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-256-finetuned-squad-seed-10 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_256_finetuned_squad_seed_2_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_256_finetuned_squad_seed_2_en.md new file mode 100644 index 00000000000000..8c862f1c57f2d8 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_256_finetuned_squad_seed_2_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_256_finetuned_squad_seed_2 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-256-finetuned-squad-seed-2` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_256_finetuned_squad_seed_2_en_4.3.0_3.0_1674214836583.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_256_finetuned_squad_seed_2_en_4.3.0_3.0_1674214836583.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_256_finetuned_squad_seed_2","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_256_finetuned_squad_seed_2","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_256_finetuned_squad_seed_2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|427.7 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-256-finetuned-squad-seed-2 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_256_finetuned_squad_seed_4_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_256_finetuned_squad_seed_4_en.md new file mode 100644 index 00000000000000..1cac84468cf56e --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_256_finetuned_squad_seed_4_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_256_finetuned_squad_seed_4 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-256-finetuned-squad-seed-4` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_256_finetuned_squad_seed_4_en_4.3.0_3.0_1674214918693.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_256_finetuned_squad_seed_4_en_4.3.0_3.0_1674214918693.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_256_finetuned_squad_seed_4","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_256_finetuned_squad_seed_4","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_256_finetuned_squad_seed_4| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|427.6 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-256-finetuned-squad-seed-4 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_256_finetuned_squad_seed_6_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_256_finetuned_squad_seed_6_en.md new file mode 100644 index 00000000000000..a2fcd76d560bbd --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_256_finetuned_squad_seed_6_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_256_finetuned_squad_seed_6 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-256-finetuned-squad-seed-6` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_256_finetuned_squad_seed_6_en_4.3.0_3.0_1674214999722.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_256_finetuned_squad_seed_6_en_4.3.0_3.0_1674214999722.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_256_finetuned_squad_seed_6","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_256_finetuned_squad_seed_6","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_256_finetuned_squad_seed_6| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|426.9 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-256-finetuned-squad-seed-6 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_256_finetuned_squad_seed_8_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_256_finetuned_squad_seed_8_en.md new file mode 100644 index 00000000000000..9cfbb8e9d38225 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_256_finetuned_squad_seed_8_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_256_finetuned_squad_seed_8 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-256-finetuned-squad-seed-8` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_256_finetuned_squad_seed_8_en_4.3.0_3.0_1674215082127.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_256_finetuned_squad_seed_8_en_4.3.0_3.0_1674215082127.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_256_finetuned_squad_seed_8","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_256_finetuned_squad_seed_8","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_256_finetuned_squad_seed_8| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|427.1 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-256-finetuned-squad-seed-8 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_32_finetuned_squad_seed_0_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_32_finetuned_squad_seed_0_en.md new file mode 100644 index 00000000000000..298d8c0da52a52 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_32_finetuned_squad_seed_0_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_32_finetuned_squad_seed_0 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-32-finetuned-squad-seed-0` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_32_finetuned_squad_seed_0_en_4.3.0_3.0_1674215168842.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_32_finetuned_squad_seed_0_en_4.3.0_3.0_1674215168842.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_32_finetuned_squad_seed_0","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_32_finetuned_squad_seed_0","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_32_finetuned_squad_seed_0| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|417.4 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-32-finetuned-squad-seed-0 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_32_finetuned_squad_seed_10_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_32_finetuned_squad_seed_10_en.md new file mode 100644 index 00000000000000..14d74d6afc289c --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_32_finetuned_squad_seed_10_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_32_finetuned_squad_seed_10 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-32-finetuned-squad-seed-10` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_32_finetuned_squad_seed_10_en_4.3.0_3.0_1674215255291.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_32_finetuned_squad_seed_10_en_4.3.0_3.0_1674215255291.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_32_finetuned_squad_seed_10","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_32_finetuned_squad_seed_10","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_32_finetuned_squad_seed_10| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|417.6 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-32-finetuned-squad-seed-10 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_512_finetuned_squad_seed_0_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_512_finetuned_squad_seed_0_en.md new file mode 100644 index 00000000000000..d584d0f603529f --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_512_finetuned_squad_seed_0_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_512_finetuned_squad_seed_0 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-512-finetuned-squad-seed-0` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_512_finetuned_squad_seed_0_en_4.3.0_3.0_1674215332011.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_512_finetuned_squad_seed_0_en_4.3.0_3.0_1674215332011.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_512_finetuned_squad_seed_0","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_512_finetuned_squad_seed_0","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_512_finetuned_squad_seed_0| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|433.1 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-512-finetuned-squad-seed-0 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_512_finetuned_squad_seed_10_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_512_finetuned_squad_seed_10_en.md new file mode 100644 index 00000000000000..a683906ceae2bb --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_512_finetuned_squad_seed_10_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_512_finetuned_squad_seed_10 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-512-finetuned-squad-seed-10` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_512_finetuned_squad_seed_10_en_4.3.0_3.0_1674215413181.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_512_finetuned_squad_seed_10_en_4.3.0_3.0_1674215413181.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_512_finetuned_squad_seed_10","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_512_finetuned_squad_seed_10","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_512_finetuned_squad_seed_10| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|433.1 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-512-finetuned-squad-seed-10 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_512_finetuned_squad_seed_2_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_512_finetuned_squad_seed_2_en.md new file mode 100644 index 00000000000000..da311da527c173 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_512_finetuned_squad_seed_2_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_512_finetuned_squad_seed_2 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-512-finetuned-squad-seed-2` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_512_finetuned_squad_seed_2_en_4.3.0_3.0_1674215491724.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_512_finetuned_squad_seed_2_en_4.3.0_3.0_1674215491724.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_512_finetuned_squad_seed_2","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_512_finetuned_squad_seed_2","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_512_finetuned_squad_seed_2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|432.9 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-512-finetuned-squad-seed-2 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_512_finetuned_squad_seed_4_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_512_finetuned_squad_seed_4_en.md new file mode 100644 index 00000000000000..dd0d2388a9089a --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_512_finetuned_squad_seed_4_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_512_finetuned_squad_seed_4 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-512-finetuned-squad-seed-4` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_512_finetuned_squad_seed_4_en_4.3.0_3.0_1674215570077.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_512_finetuned_squad_seed_4_en_4.3.0_3.0_1674215570077.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_512_finetuned_squad_seed_4","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_512_finetuned_squad_seed_4","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_512_finetuned_squad_seed_4| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|432.9 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-512-finetuned-squad-seed-4 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_512_finetuned_squad_seed_6_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_512_finetuned_squad_seed_6_en.md new file mode 100644 index 00000000000000..47339ad99dc2ed --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_512_finetuned_squad_seed_6_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_512_finetuned_squad_seed_6 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-512-finetuned-squad-seed-6` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_512_finetuned_squad_seed_6_en_4.3.0_3.0_1674215648700.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_512_finetuned_squad_seed_6_en_4.3.0_3.0_1674215648700.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_512_finetuned_squad_seed_6","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_512_finetuned_squad_seed_6","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_512_finetuned_squad_seed_6| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|432.9 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-512-finetuned-squad-seed-6 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_512_finetuned_squad_seed_8_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_512_finetuned_squad_seed_8_en.md new file mode 100644 index 00000000000000..28d4609f57abae --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_512_finetuned_squad_seed_8_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_512_finetuned_squad_seed_8 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-512-finetuned-squad-seed-8` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_512_finetuned_squad_seed_8_en_4.3.0_3.0_1674215724845.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_512_finetuned_squad_seed_8_en_4.3.0_3.0_1674215724845.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_512_finetuned_squad_seed_8","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_512_finetuned_squad_seed_8","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_512_finetuned_squad_seed_8| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|432.8 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-512-finetuned-squad-seed-8 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_64_finetuned_squad_seed_0_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_64_finetuned_squad_seed_0_en.md new file mode 100644 index 00000000000000..bea53ab90d5d9b --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_64_finetuned_squad_seed_0_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_64_finetuned_squad_seed_0 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-64-finetuned-squad-seed-0` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_64_finetuned_squad_seed_0_en_4.3.0_3.0_1674215810378.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_64_finetuned_squad_seed_0_en_4.3.0_3.0_1674215810378.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_64_finetuned_squad_seed_0","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_64_finetuned_squad_seed_0","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_64_finetuned_squad_seed_0| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|419.4 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-64-finetuned-squad-seed-0 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_64_finetuned_squad_seed_10_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_64_finetuned_squad_seed_10_en.md new file mode 100644 index 00000000000000..37ddba6d018b63 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_64_finetuned_squad_seed_10_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_64_finetuned_squad_seed_10 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-64-finetuned-squad-seed-10` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_64_finetuned_squad_seed_10_en_4.3.0_3.0_1674215893984.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_64_finetuned_squad_seed_10_en_4.3.0_3.0_1674215893984.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_64_finetuned_squad_seed_10","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_64_finetuned_squad_seed_10","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_64_finetuned_squad_seed_10| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|419.8 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-64-finetuned-squad-seed-10 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_64_finetuned_squad_seed_2_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_64_finetuned_squad_seed_2_en.md new file mode 100644 index 00000000000000..18dd0343abdb31 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_64_finetuned_squad_seed_2_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_64_finetuned_squad_seed_2 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-64-finetuned-squad-seed-2` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_64_finetuned_squad_seed_2_en_4.3.0_3.0_1674215981311.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_64_finetuned_squad_seed_2_en_4.3.0_3.0_1674215981311.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_64_finetuned_squad_seed_2","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_64_finetuned_squad_seed_2","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_64_finetuned_squad_seed_2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|419.9 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-64-finetuned-squad-seed-2 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_64_finetuned_squad_seed_4_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_64_finetuned_squad_seed_4_en.md new file mode 100644 index 00000000000000..12dba3f0711c06 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_64_finetuned_squad_seed_4_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_64_finetuned_squad_seed_4 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-64-finetuned-squad-seed-4` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_64_finetuned_squad_seed_4_en_4.3.0_3.0_1674216065319.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_64_finetuned_squad_seed_4_en_4.3.0_3.0_1674216065319.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_64_finetuned_squad_seed_4","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_64_finetuned_squad_seed_4","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_64_finetuned_squad_seed_4| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|419.7 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-64-finetuned-squad-seed-4 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_64_finetuned_squad_seed_6_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_64_finetuned_squad_seed_6_en.md new file mode 100644 index 00000000000000..e5d0696f0e2d95 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_64_finetuned_squad_seed_6_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_64_finetuned_squad_seed_6 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-64-finetuned-squad-seed-6` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_64_finetuned_squad_seed_6_en_4.3.0_3.0_1674216152630.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_64_finetuned_squad_seed_6_en_4.3.0_3.0_1674216152630.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_64_finetuned_squad_seed_6","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_64_finetuned_squad_seed_6","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_64_finetuned_squad_seed_6| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|419.5 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-64-finetuned-squad-seed-6 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_64_finetuned_squad_seed_8_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_64_finetuned_squad_seed_8_en.md new file mode 100644 index 00000000000000..9594ee2a3b042b --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_few_shot_k_64_finetuned_squad_seed_8_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_few_shot_k_64_finetuned_squad_seed_8 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-few-shot-k-64-finetuned-squad-seed-8` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_64_finetuned_squad_seed_8_en_4.3.0_3.0_1674216234526.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_few_shot_k_64_finetuned_squad_seed_8_en_4.3.0_3.0_1674216234526.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_64_finetuned_squad_seed_8","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_few_shot_k_64_finetuned_squad_seed_8","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_few_shot_k_64_finetuned_squad_seed_8| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|419.6 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-few-shot-k-64-finetuned-squad-seed-8 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_filtered_cuad_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_filtered_cuad_en.md new file mode 100644 index 00000000000000..7a8c6ae593ca0e --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_filtered_cuad_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from alex-apostolo) +author: John Snow Labs +name: roberta_qa_base_filtered_cuad +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-filtered-cuad` is a English model originally trained by `alex-apostolo`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_filtered_cuad_en_4.3.0_3.0_1674216293189.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_filtered_cuad_en_4.3.0_3.0_1674216293189.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_filtered_cuad","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_filtered_cuad","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_filtered_cuad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|454.4 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/alex-apostolo/roberta-base-filtered-cuad \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_cuad_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_cuad_en.md new file mode 100644 index 00000000000000..829fb65ecdbdc0 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_cuad_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from Gam) +author: John Snow Labs +name: roberta_qa_base_finetuned_cuad +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-finetuned-cuad` is a English model originally trained by `Gam`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_cuad_en_4.3.0_3.0_1674216413698.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_cuad_en_4.3.0_3.0_1674216413698.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_cuad","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_cuad","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_finetuned_cuad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|451.7 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Gam/roberta-base-finetuned-cuad \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_cuad_gam_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_cuad_gam_en.md new file mode 100644 index 00000000000000..5f4985f174ee0d --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_cuad_gam_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from Gam) +author: John Snow Labs +name: roberta_qa_base_finetuned_cuad_gam +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-finetuned-cuad-gam` is a English model originally trained by `Gam`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_cuad_gam_en_4.3.0_3.0_1674216483026.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_cuad_gam_en_4.3.0_3.0_1674216483026.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_cuad_gam","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_cuad_gam","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_finetuned_cuad_gam| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|450.3 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Gam/roberta-base-finetuned-cuad-gam \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_deletion_squad_10_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_deletion_squad_10_en.md new file mode 100644 index 00000000000000..2778a02dd3f7c3 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_deletion_squad_10_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from huxxx657) +author: John Snow Labs +name: roberta_qa_base_finetuned_deletion_squad_10 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-finetuned-deletion-squad-10` is a English model originally trained by `huxxx657`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_deletion_squad_10_en_4.3.0_3.0_1674216541297.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_deletion_squad_10_en_4.3.0_3.0_1674216541297.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_deletion_squad_10","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_deletion_squad_10","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_finetuned_deletion_squad_10| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.2 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/huxxx657/roberta-base-finetuned-deletion-squad-10 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_deletion_squad_15_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_deletion_squad_15_en.md new file mode 100644 index 00000000000000..90038bccc8316e --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_deletion_squad_15_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from huxxx657) +author: John Snow Labs +name: roberta_qa_base_finetuned_deletion_squad_15 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-finetuned-deletion-squad-15` is a English model originally trained by `huxxx657`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_deletion_squad_15_en_4.3.0_3.0_1674216599443.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_deletion_squad_15_en_4.3.0_3.0_1674216599443.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_deletion_squad_15","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_deletion_squad_15","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_finetuned_deletion_squad_15| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.3 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/huxxx657/roberta-base-finetuned-deletion-squad-15 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_en.md new file mode 100644 index 00000000000000..60993820c4082c --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from 123tarunanand) +author: John Snow Labs +name: roberta_qa_base_finetuned +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-finetuned` is a English model originally trained by `123tarunanand`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_en_4.3.0_3.0_1674216346492.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_en_4.3.0_3.0_1674216346492.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_finetuned| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.4 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/123tarunanand/roberta-base-finetuned \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_hotpot_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_hotpot_en.md new file mode 100644 index 00000000000000..469d5d33532165 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_hotpot_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from clevrly) +author: John Snow Labs +name: roberta_qa_base_finetuned_hotpot +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-finetuned-hotpot_qa` is a English model originally trained by `clevrly`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_hotpot_en_4.3.0_3.0_1674216656084.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_hotpot_en_4.3.0_3.0_1674216656084.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_hotpot","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_hotpot","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_finetuned_hotpot| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.6 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/clevrly/roberta-base-finetuned-hotpot_qa \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_scrambled_squad_10_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_scrambled_squad_10_en.md new file mode 100644 index 00000000000000..afd9c5cd8f9204 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_scrambled_squad_10_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from huxxx657) +author: John Snow Labs +name: roberta_qa_base_finetuned_scrambled_squad_10 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-finetuned-scrambled-squad-10` is a English model originally trained by `huxxx657`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_scrambled_squad_10_en_4.3.0_3.0_1674216712953.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_scrambled_squad_10_en_4.3.0_3.0_1674216712953.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_scrambled_squad_10","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_scrambled_squad_10","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_finetuned_scrambled_squad_10| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.0 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/huxxx657/roberta-base-finetuned-scrambled-squad-10 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_scrambled_squad_10_new_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_scrambled_squad_10_new_en.md new file mode 100644 index 00000000000000..ab82c9ef59f9c2 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_scrambled_squad_10_new_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from huxxx657) +author: John Snow Labs +name: roberta_qa_base_finetuned_scrambled_squad_10_new +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-finetuned-scrambled-squad-10-new` is a English model originally trained by `huxxx657`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_scrambled_squad_10_new_en_4.3.0_3.0_1674216770257.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_scrambled_squad_10_new_en_4.3.0_3.0_1674216770257.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_scrambled_squad_10_new","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_scrambled_squad_10_new","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_finetuned_scrambled_squad_10_new| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.4 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/huxxx657/roberta-base-finetuned-scrambled-squad-10-new \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_scrambled_squad_15_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_scrambled_squad_15_en.md new file mode 100644 index 00000000000000..dffae3a3e4eb2c --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_scrambled_squad_15_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from huxxx657) +author: John Snow Labs +name: roberta_qa_base_finetuned_scrambled_squad_15 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-finetuned-scrambled-squad-15` is a English model originally trained by `huxxx657`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_scrambled_squad_15_en_4.3.0_3.0_1674216826395.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_scrambled_squad_15_en_4.3.0_3.0_1674216826395.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_scrambled_squad_15","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_scrambled_squad_15","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_finetuned_scrambled_squad_15| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.3 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/huxxx657/roberta-base-finetuned-scrambled-squad-15 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_scrambled_squad_15_new_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_scrambled_squad_15_new_en.md new file mode 100644 index 00000000000000..304b87a1c8ae32 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_scrambled_squad_15_new_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from huxxx657) +author: John Snow Labs +name: roberta_qa_base_finetuned_scrambled_squad_15_new +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-finetuned-scrambled-squad-15-new` is a English model originally trained by `huxxx657`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_scrambled_squad_15_new_en_4.3.0_3.0_1674216883682.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_scrambled_squad_15_new_en_4.3.0_3.0_1674216883682.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_scrambled_squad_15_new","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_scrambled_squad_15_new","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_finetuned_scrambled_squad_15_new| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.5 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/huxxx657/roberta-base-finetuned-scrambled-squad-15-new \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_scrambled_squad_5_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_scrambled_squad_5_en.md new file mode 100644 index 00000000000000..13b4dc9572037a --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_scrambled_squad_5_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from huxxx657) +author: John Snow Labs +name: roberta_qa_base_finetuned_scrambled_squad_5 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-finetuned-scrambled-squad-5` is a English model originally trained by `huxxx657`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_scrambled_squad_5_en_4.3.0_3.0_1674216944002.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_scrambled_squad_5_en_4.3.0_3.0_1674216944002.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_scrambled_squad_5","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_scrambled_squad_5","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_finetuned_scrambled_squad_5| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.0 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/huxxx657/roberta-base-finetuned-scrambled-squad-5 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_scrambled_squad_5_new_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_scrambled_squad_5_new_en.md new file mode 100644 index 00000000000000..29edb79e7dbcc1 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_scrambled_squad_5_new_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from huxxx657) +author: John Snow Labs +name: roberta_qa_base_finetuned_scrambled_squad_5_new +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-finetuned-scrambled-squad-5-new` is a English model originally trained by `huxxx657`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_scrambled_squad_5_new_en_4.3.0_3.0_1674216999446.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_scrambled_squad_5_new_en_4.3.0_3.0_1674216999446.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_scrambled_squad_5_new","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_scrambled_squad_5_new","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_finetuned_scrambled_squad_5_new| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.4 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/huxxx657/roberta-base-finetuned-scrambled-squad-5-new \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_squad2_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_squad2_en.md new file mode 100644 index 00000000000000..09d9f6c0fcb12e --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_squad2_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from mvonwyl) +author: John Snow Labs +name: roberta_qa_base_finetuned_squad2 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-finetuned-squad2` is a English model originally trained by `mvonwyl`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_squad2_en_4.3.0_3.0_1674217768093.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_squad2_en_4.3.0_3.0_1674217768093.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_squad2","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_squad2","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_finetuned_squad2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.4 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/mvonwyl/roberta-base-finetuned-squad2 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_squad2_lwt_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_squad2_lwt_en.md new file mode 100644 index 00000000000000..650856949fae73 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_squad2_lwt_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from 21iridescent) +author: John Snow Labs +name: roberta_qa_base_finetuned_squad2_lwt +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `distilroberta-base-finetuned-squad2-lwt` is a English model originally trained by `21iridescent`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_squad2_lwt_en_4.3.0_3.0_1674210398377.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_squad2_lwt_en_4.3.0_3.0_1674210398377.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_squad2_lwt","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_squad2_lwt","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_finetuned_squad2_lwt| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|307.1 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/21iridescent/distilroberta-base-finetuned-squad2-lwt \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_squad_1_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_squad_1_en.md new file mode 100644 index 00000000000000..0e49fe372451ba --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_squad_1_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from huxxx657) +author: John Snow Labs +name: roberta_qa_base_finetuned_squad_1 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-finetuned-squad-1` is a English model originally trained by `huxxx657`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_squad_1_en_4.3.0_3.0_1674217536049.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_squad_1_en_4.3.0_3.0_1674217536049.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_squad_1","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_squad_1","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_finetuned_squad_1| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.1 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/huxxx657/roberta-base-finetuned-squad-1 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_squad_2_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_squad_2_en.md new file mode 100644 index 00000000000000..30e388f2f7ab86 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_squad_2_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from huxxx657) +author: John Snow Labs +name: roberta_qa_base_finetuned_squad_2 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-finetuned-squad-2` is a English model originally trained by `huxxx657`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_squad_2_en_4.3.0_3.0_1674217595105.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_squad_2_en_4.3.0_3.0_1674217595105.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_squad_2","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_squad_2","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_finetuned_squad_2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|438.0 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/huxxx657/roberta-base-finetuned-squad-2 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_squad_3_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_squad_3_en.md new file mode 100644 index 00000000000000..2f30eaa3decb08 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_squad_3_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from huxxx657) +author: John Snow Labs +name: roberta_qa_base_finetuned_squad_3 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-finetuned-squad-3` is a English model originally trained by `huxxx657`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_squad_3_en_4.3.0_3.0_1674217654479.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_squad_3_en_4.3.0_3.0_1674217654479.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_squad_3","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_squad_3","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_finetuned_squad_3| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.2 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/huxxx657/roberta-base-finetuned-squad-3 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_squad_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_squad_en.md new file mode 100644 index 00000000000000..e23debf2901bb6 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_squad_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from victorlee071200) +author: John Snow Labs +name: roberta_qa_base_finetuned_squad +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `distilroberta-base-finetuned-squad` is a English model originally trained by `victorlee071200`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_squad_en_4.3.0_3.0_1674210358228.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_squad_en_4.3.0_3.0_1674210358228.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_squad","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_squad","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_finetuned_squad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|307.1 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/victorlee071200/distilroberta-base-finetuned-squad \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_squad_r3f_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_squad_r3f_en.md new file mode 100644 index 00000000000000..bb875c0b682d4b --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_squad_r3f_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_finetuned_squad_r3f +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-finetuned-squad-r3f` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_squad_r3f_en_4.3.0_3.0_1674217712089.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_squad_r3f_en_4.3.0_3.0_1674217712089.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_squad_r3f","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_squad_r3f","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_finetuned_squad_r3f| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.2 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-base-finetuned-squad-r3f \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_squad_v2_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_squad_v2_en.md new file mode 100644 index 00000000000000..ce227b6e13c3a5 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_finetuned_squad_v2_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from victorlee071200) +author: John Snow Labs +name: roberta_qa_base_finetuned_squad_v2 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `distilroberta-base-finetuned-squad_v2` is a English model originally trained by `victorlee071200`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_squad_v2_en_4.3.0_3.0_1674210438571.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_finetuned_squad_v2_en_4.3.0_3.0_1674210438571.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_squad_v2","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_finetuned_squad_v2","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_finetuned_squad_v2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|307.1 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/victorlee071200/distilroberta-base-finetuned-squad_v2 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_ft_esg_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_ft_esg_en.md new file mode 100644 index 00000000000000..d4339e41fc63c1 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_ft_esg_en.md @@ -0,0 +1,89 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from Ayushb) +author: John Snow Labs +name: roberta_qa_base_ft_esg +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-ft-esg` is a English model originally trained by `Ayushb`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_ft_esg_en_4.3.0_3.0_1674217855538.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_ft_esg_en_4.3.0_3.0_1674217855538.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_ft_esg","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_ft_esg","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_ft_esg| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|416.4 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Ayushb/roberta-base-ft-esg +- https://www.github.com/Ayush1702 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_qna_squad2_trained_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_qna_squad2_trained_en.md new file mode 100644 index 00000000000000..7e4f3bc91a6c88 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_qna_squad2_trained_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from Shappey) +author: John Snow Labs +name: roberta_qa_base_qna_squad2_trained +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-QnA-squad2-trained` is a English model originally trained by `Shappey`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_qna_squad2_trained_en_4.3.0_3.0_1674212572106.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_qna_squad2_trained_en_4.3.0_3.0_1674212572106.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_qna_squad2_trained","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_qna_squad2_trained","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_qna_squad2_trained| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|456.8 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Shappey/roberta-base-QnA-squad2-trained \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_spanish_squades_becas1_es.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_spanish_squades_becas1_es.md new file mode 100644 index 00000000000000..47c7a4614068ae --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_spanish_squades_becas1_es.md @@ -0,0 +1,88 @@ +--- +layout: model +title: Spanish RobertaForQuestionAnswering Base Cased model (from Evelyn18) +author: John Snow Labs +name: roberta_qa_base_spanish_squades_becas1 +date: 2023-01-20 +tags: [es, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: es +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-spanish-squades-becas1` is a Spanish model originally trained by `Evelyn18`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_spanish_squades_becas1_es_4.3.0_3.0_1674217912605.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_spanish_squades_becas1_es_4.3.0_3.0_1674217912605.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_spanish_squades_becas1","es")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_spanish_squades_becas1","es") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_spanish_squades_becas1| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|es| +|Size:|460.0 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Evelyn18/roberta-base-spanish-squades-becas1 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_spanish_squades_becasincentivos1_es.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_spanish_squades_becasincentivos1_es.md new file mode 100644 index 00000000000000..2e12e116c2e79b --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_spanish_squades_becasincentivos1_es.md @@ -0,0 +1,88 @@ +--- +layout: model +title: Spanish RobertaForQuestionAnswering Base Cased model (from Evelyn18) +author: John Snow Labs +name: roberta_qa_base_spanish_squades_becasincentivos1 +date: 2023-01-20 +tags: [es, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: es +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-spanish-squades-becasIncentivos1` is a Spanish model originally trained by `Evelyn18`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_spanish_squades_becasincentivos1_es_4.3.0_3.0_1674217969589.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_spanish_squades_becasincentivos1_es_4.3.0_3.0_1674217969589.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_spanish_squades_becasincentivos1","es")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_spanish_squades_becasincentivos1","es") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_spanish_squades_becasincentivos1| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|es| +|Size:|459.9 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Evelyn18/roberta-base-spanish-squades-becasIncentivos1 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_spanish_squades_becasincentivos2_es.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_spanish_squades_becasincentivos2_es.md new file mode 100644 index 00000000000000..a757d3d6954b9c --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_spanish_squades_becasincentivos2_es.md @@ -0,0 +1,88 @@ +--- +layout: model +title: Spanish RobertaForQuestionAnswering Base Cased model (from Evelyn18) +author: John Snow Labs +name: roberta_qa_base_spanish_squades_becasincentivos2 +date: 2023-01-20 +tags: [es, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: es +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-spanish-squades-becasIncentivos2` is a Spanish model originally trained by `Evelyn18`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_spanish_squades_becasincentivos2_es_4.3.0_3.0_1674218030841.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_spanish_squades_becasincentivos2_es_4.3.0_3.0_1674218030841.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_spanish_squades_becasincentivos2","es")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_spanish_squades_becasincentivos2","es") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_spanish_squades_becasincentivos2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|es| +|Size:|459.8 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Evelyn18/roberta-base-spanish-squades-becasIncentivos2 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_spanish_squades_becasincentivos3_es.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_spanish_squades_becasincentivos3_es.md new file mode 100644 index 00000000000000..bebf584cffc06a --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_spanish_squades_becasincentivos3_es.md @@ -0,0 +1,88 @@ +--- +layout: model +title: Spanish RobertaForQuestionAnswering Base Cased model (from Evelyn18) +author: John Snow Labs +name: roberta_qa_base_spanish_squades_becasincentivos3 +date: 2023-01-20 +tags: [es, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: es +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-spanish-squades-becasIncentivos3` is a Spanish model originally trained by `Evelyn18`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_spanish_squades_becasincentivos3_es_4.3.0_3.0_1674218087235.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_spanish_squades_becasincentivos3_es_4.3.0_3.0_1674218087235.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_spanish_squades_becasincentivos3","es")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_spanish_squades_becasincentivos3","es") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_spanish_squades_becasincentivos3| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|es| +|Size:|459.8 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Evelyn18/roberta-base-spanish-squades-becasIncentivos3 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_spanish_squades_becasincentivos4_es.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_spanish_squades_becasincentivos4_es.md new file mode 100644 index 00000000000000..cf4e8d32272caa --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_spanish_squades_becasincentivos4_es.md @@ -0,0 +1,88 @@ +--- +layout: model +title: Spanish RobertaForQuestionAnswering Base Cased model (from Evelyn18) +author: John Snow Labs +name: roberta_qa_base_spanish_squades_becasincentivos4 +date: 2023-01-20 +tags: [es, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: es +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-spanish-squades-becasIncentivos4` is a Spanish model originally trained by `Evelyn18`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_spanish_squades_becasincentivos4_es_4.3.0_3.0_1674218146985.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_spanish_squades_becasincentivos4_es_4.3.0_3.0_1674218146985.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_spanish_squades_becasincentivos4","es")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_spanish_squades_becasincentivos4","es") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_spanish_squades_becasincentivos4| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|es| +|Size:|459.8 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Evelyn18/roberta-base-spanish-squades-becasIncentivos4 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_spanish_squades_becasincentivos6_es.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_spanish_squades_becasincentivos6_es.md new file mode 100644 index 00000000000000..8c094f4baf3bbd --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_spanish_squades_becasincentivos6_es.md @@ -0,0 +1,88 @@ +--- +layout: model +title: Spanish RobertaForQuestionAnswering Base Cased model (from Evelyn18) +author: John Snow Labs +name: roberta_qa_base_spanish_squades_becasincentivos6 +date: 2023-01-20 +tags: [es, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: es +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-spanish-squades-becasIncentivos6` is a Spanish model originally trained by `Evelyn18`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_spanish_squades_becasincentivos6_es_4.3.0_3.0_1674218206378.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_spanish_squades_becasincentivos6_es_4.3.0_3.0_1674218206378.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_spanish_squades_becasincentivos6","es")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_spanish_squades_becasincentivos6","es") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_spanish_squades_becasincentivos6| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|es| +|Size:|459.9 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Evelyn18/roberta-base-spanish-squades-becasIncentivos6 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_spanish_squades_becasv3_es.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_spanish_squades_becasv3_es.md new file mode 100644 index 00000000000000..abb58eb926e47c --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_spanish_squades_becasv3_es.md @@ -0,0 +1,88 @@ +--- +layout: model +title: Spanish RobertaForQuestionAnswering Base Cased model (from Evelyn18) +author: John Snow Labs +name: roberta_qa_base_spanish_squades_becasv3 +date: 2023-01-20 +tags: [es, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: es +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-spanish-squades-becasv3` is a Spanish model originally trained by `Evelyn18`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_spanish_squades_becasv3_es_4.3.0_3.0_1674218266592.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_spanish_squades_becasv3_es_4.3.0_3.0_1674218266592.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_spanish_squades_becasv3","es")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_spanish_squades_becasv3","es") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_spanish_squades_becasv3| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|es| +|Size:|459.9 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Evelyn18/roberta-base-spanish-squades-becasv3 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_spanish_squades_modelo1_es.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_spanish_squades_modelo1_es.md new file mode 100644 index 00000000000000..a506561a8c671e --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_spanish_squades_modelo1_es.md @@ -0,0 +1,88 @@ +--- +layout: model +title: Spanish RobertaForQuestionAnswering Base Cased model (from Evelyn18) +author: John Snow Labs +name: roberta_qa_base_spanish_squades_modelo1 +date: 2023-01-20 +tags: [es, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: es +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-spanish-squades-modelo1` is a Spanish model originally trained by `Evelyn18`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_spanish_squades_modelo1_es_4.3.0_3.0_1674218442505.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_spanish_squades_modelo1_es_4.3.0_3.0_1674218442505.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_spanish_squades_modelo1","es")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_spanish_squades_modelo1","es") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_spanish_squades_modelo1| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|es| +|Size:|460.0 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Evelyn18/roberta-base-spanish-squades-modelo1 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_spanish_squades_modelo2_es.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_spanish_squades_modelo2_es.md new file mode 100644 index 00000000000000..7d5336018fd34b --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_spanish_squades_modelo2_es.md @@ -0,0 +1,88 @@ +--- +layout: model +title: Spanish RobertaForQuestionAnswering Base Cased model (from Evelyn18) +author: John Snow Labs +name: roberta_qa_base_spanish_squades_modelo2 +date: 2023-01-20 +tags: [es, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: es +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-spanish-squades-modelo2` is a Spanish model originally trained by `Evelyn18`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_spanish_squades_modelo2_es_4.3.0_3.0_1674218501375.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_spanish_squades_modelo2_es_4.3.0_3.0_1674218501375.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_spanish_squades_modelo2","es")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_spanish_squades_modelo2","es") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_spanish_squades_modelo2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|es| +|Size:|459.9 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Evelyn18/roberta-base-spanish-squades-modelo2 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_spanish_squades_modelo_v1_es.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_spanish_squades_modelo_v1_es.md new file mode 100644 index 00000000000000..eff5a29b02c9a6 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_spanish_squades_modelo_v1_es.md @@ -0,0 +1,88 @@ +--- +layout: model +title: Spanish RobertaForQuestionAnswering Base Cased model (from Evelyn18) +author: John Snow Labs +name: roberta_qa_base_spanish_squades_modelo_v1 +date: 2023-01-20 +tags: [es, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: es +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-spanish-squades-modelo-robertav1` is a Spanish model originally trained by `Evelyn18`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_spanish_squades_modelo_v1_es_4.3.0_3.0_1674218326080.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_spanish_squades_modelo_v1_es_4.3.0_3.0_1674218326080.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_spanish_squades_modelo_v1","es")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_spanish_squades_modelo_v1","es") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_spanish_squades_modelo_v1| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|es| +|Size:|460.0 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Evelyn18/roberta-base-spanish-squades-modelo-robertav1 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_spanish_squades_modelo_v1b3_es.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_spanish_squades_modelo_v1b3_es.md new file mode 100644 index 00000000000000..11f324402fbc44 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_spanish_squades_modelo_v1b3_es.md @@ -0,0 +1,88 @@ +--- +layout: model +title: Spanish RobertaForQuestionAnswering Base Cased model (from Evelyn18) +author: John Snow Labs +name: roberta_qa_base_spanish_squades_modelo_v1b3 +date: 2023-01-20 +tags: [es, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: es +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-spanish-squades-modelo-robertav1b3` is a Spanish model originally trained by `Evelyn18`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_spanish_squades_modelo_v1b3_es_4.3.0_3.0_1674218384530.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_spanish_squades_modelo_v1b3_es_4.3.0_3.0_1674218384530.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_spanish_squades_modelo_v1b3","es")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_spanish_squades_modelo_v1b3","es") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_spanish_squades_modelo_v1b3| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|es| +|Size:|460.0 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Evelyn18/roberta-base-spanish-squades-modelo-robertav1b3 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_spanish_squades_v2_es.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_spanish_squades_v2_es.md new file mode 100644 index 00000000000000..b3e80c953fb62b --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_spanish_squades_v2_es.md @@ -0,0 +1,88 @@ +--- +layout: model +title: Spanish RobertaForQuestionAnswering Base Cased model (from Evelyn18) +author: John Snow Labs +name: roberta_qa_base_spanish_squades_v2 +date: 2023-01-20 +tags: [es, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: es +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-spanish-squades-robertav2` is a Spanish model originally trained by `Evelyn18`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_spanish_squades_v2_es_4.3.0_3.0_1674218559260.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_spanish_squades_v2_es_4.3.0_3.0_1674218559260.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_spanish_squades_v2","es")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_spanish_squades_v2","es") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_spanish_squades_v2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|es| +|Size:|460.0 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Evelyn18/roberta-base-spanish-squades-robertav2 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad2.0_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad2.0_en.md new file mode 100644 index 00000000000000..4d14c8d54b221d --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad2.0_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_base_squad2.0 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base_squad2.0` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_squad2.0_en_4.3.0_3.0_1674219848563.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_squad2.0_en_4.3.0_3.0_1674219848563.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_squad2.0","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_squad2.0","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_squad2.0| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|460.4 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/roberta-base_squad2.0 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad2_boolq_baseline_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad2_boolq_baseline_en.md new file mode 100644 index 00000000000000..57374a46dd6717 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad2_boolq_baseline_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from shahrukhx01) +author: John Snow Labs +name: roberta_qa_base_squad2_boolq_baseline +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-squad2-boolq-baseline` is a English model originally trained by `shahrukhx01`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_squad2_boolq_baseline_en_4.3.0_3.0_1674219076102.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_squad2_boolq_baseline_en_4.3.0_3.0_1674219076102.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_squad2_boolq_baseline","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_squad2_boolq_baseline","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_squad2_boolq_baseline| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.3 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/shahrukhx01/roberta-base-squad2-boolq-baseline \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad2_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad2_en.md new file mode 100644 index 00000000000000..c756103d017385 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad2_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from twmkn9) +author: John Snow Labs +name: roberta_qa_base_squad2 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `distilroberta-base-squad2` is a English model originally trained by `twmkn9`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_squad2_en_4.3.0_3.0_1674210478798.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_squad2_en_4.3.0_3.0_1674210478798.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_squad2","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_squad2","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_squad2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|307.1 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/twmkn9/distilroberta-base-squad2 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad2_finetuned_sel_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad2_finetuned_sel_en.md new file mode 100644 index 00000000000000..2a26d236171a8c --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad2_finetuned_sel_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from Teepika) +author: John Snow Labs +name: roberta_qa_base_squad2_finetuned_sel +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-squad2-finetuned-selqa` is a English model originally trained by `Teepika`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_squad2_finetuned_sel_en_4.3.0_3.0_1674219133811.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_squad2_finetuned_sel_en_4.3.0_3.0_1674219133811.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_squad2_finetuned_sel","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_squad2_finetuned_sel","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_squad2_finetuned_sel| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.8 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Teepika/roberta-base-squad2-finetuned-selqa \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad2_finetuned_squad2_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad2_finetuned_squad2_en.md new file mode 100644 index 00000000000000..8e298a6909fd4d --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad2_finetuned_squad2_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from Marscen) +author: John Snow Labs +name: roberta_qa_base_squad2_finetuned_squad2 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-squad2-finetuned-squad2` is a English model originally trained by `Marscen`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_squad2_finetuned_squad2_en_4.3.0_3.0_1674219552956.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_squad2_finetuned_squad2_en_4.3.0_3.0_1674219552956.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_squad2_finetuned_squad2","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_squad2_finetuned_squad2","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_squad2_finetuned_squad2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.3 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Marscen/roberta-base-squad2-finetuned-squad2 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad2_finetuned_visquad_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad2_finetuned_visquad_en.md new file mode 100644 index 00000000000000..03824fb45b7ebc --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad2_finetuned_visquad_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from khoanvm) +author: John Snow Labs +name: roberta_qa_base_squad2_finetuned_visquad +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-squad2-finetuned-visquad` is a English model originally trained by `khoanvm`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_squad2_finetuned_visquad_en_4.3.0_3.0_1674219613502.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_squad2_finetuned_visquad_en_4.3.0_3.0_1674219613502.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_squad2_finetuned_visquad","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_squad2_finetuned_visquad","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_squad2_finetuned_visquad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.3 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/khoanvm/roberta-base-squad2-finetuned-visquad \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad2_nq_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad2_nq_en.md new file mode 100644 index 00000000000000..c869fed5d040a9 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad2_nq_en.md @@ -0,0 +1,89 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from nlpconnect) +author: John Snow Labs +name: roberta_qa_base_squad2_nq +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-squad2-nq` is a English model originally trained by `nlpconnect`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_squad2_nq_en_4.3.0_3.0_1674219670141.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_squad2_nq_en_4.3.0_3.0_1674219670141.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_squad2_nq","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_squad2_nq","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_squad2_nq| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.3 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/nlpconnect/roberta-base-squad2-nq +- https://paperswithcode.com/sota?task=Question+Answering&dataset=squad_v2 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad2_ta_qna_10e_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad2_ta_qna_10e_en.md new file mode 100644 index 00000000000000..dc56b97a8f652c --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad2_ta_qna_10e_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from venkateshdas) +author: John Snow Labs +name: roberta_qa_base_squad2_ta_qna_10e +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-squad2-ta-qna-roberta10e` is a English model originally trained by `venkateshdas`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_squad2_ta_qna_10e_en_4.3.0_3.0_1674219728754.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_squad2_ta_qna_10e_en_4.3.0_3.0_1674219728754.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_squad2_ta_qna_10e","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_squad2_ta_qna_10e","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_squad2_ta_qna_10e| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.3 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/venkateshdas/roberta-base-squad2-ta-qna-roberta10e \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad2_ta_qna_3e_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad2_ta_qna_3e_en.md new file mode 100644 index 00000000000000..4370a4407c6a77 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad2_ta_qna_3e_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from venkateshdas) +author: John Snow Labs +name: roberta_qa_base_squad2_ta_qna_3e +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-squad2-ta-qna-roberta3e` is a English model originally trained by `venkateshdas`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_squad2_ta_qna_3e_en_4.3.0_3.0_1674219788427.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_squad2_ta_qna_3e_en_4.3.0_3.0_1674219788427.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_squad2_ta_qna_3e","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_squad2_ta_qna_3e","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_squad2_ta_qna_3e| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.3 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/venkateshdas/roberta-base-squad2-ta-qna-roberta3e \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad_en.md new file mode 100644 index 00000000000000..b94aa125e14fab --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from Palak) +author: John Snow Labs +name: roberta_qa_base_squad +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `distilroberta-base_squad` is a English model originally trained by `Palak`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_squad_en_4.3.0_3.0_1674210601142.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_squad_en_4.3.0_3.0_1674210601142.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_squad","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_squad","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_squad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|307.1 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Palak/distilroberta-base_squad \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad_finetuned_on_runaways_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad_finetuned_on_runaways_en.md new file mode 100644 index 00000000000000..99d0c69bb25e18 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad_finetuned_on_runaways_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from Nadav) +author: John Snow Labs +name: roberta_qa_base_squad_finetuned_on_runaways +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-squad-finetuned-on-runaways-en` is a English model originally trained by `Nadav`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_squad_finetuned_on_runaways_en_4.3.0_3.0_1674218728000.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_squad_finetuned_on_runaways_en_4.3.0_3.0_1674218728000.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_squad_finetuned_on_runaways","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_squad_finetuned_on_runaways","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_squad_finetuned_on_runaways| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|467.0 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Nadav/roberta-base-squad-finetuned-on-runaways-en \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad_finetuned_squad_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad_finetuned_squad_en.md new file mode 100644 index 00000000000000..2e9f1b9b509add --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad_finetuned_squad_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from Swty) +author: John Snow Labs +name: roberta_qa_base_squad_finetuned_squad +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-squad-finetuned-squad` is a English model originally trained by `Swty`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_squad_finetuned_squad_en_4.3.0_3.0_1674218785890.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_squad_finetuned_squad_en_4.3.0_3.0_1674218785890.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_squad_finetuned_squad","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_squad_finetuned_squad","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_squad_finetuned_squad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.3 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Swty/roberta-base-squad-finetuned-squad \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad_nl.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad_nl.md new file mode 100644 index 00000000000000..2674babc920c13 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_squad_nl.md @@ -0,0 +1,88 @@ +--- +layout: model +title: Dutch RobertaForQuestionAnswering Base Cased model (from Nadav) +author: John Snow Labs +name: roberta_qa_base_squad +date: 2023-01-20 +tags: [nl, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: nl +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-squad-nl` is a Dutch model originally trained by `Nadav`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_squad_nl_4.3.0_3.0_1674218840970.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_squad_nl_4.3.0_3.0_1674218840970.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_squad","nl")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_squad","nl") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_squad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|nl| +|Size:|436.5 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Nadav/roberta-base-squad-nl \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_super_1_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_super_1_en.md new file mode 100644 index 00000000000000..5cdc94c62932e3 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_super_1_en.md @@ -0,0 +1,89 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from nbroad) +author: John Snow Labs +name: roberta_qa_base_super_1 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `rob-base-superqa1` is a English model originally trained by `nbroad`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_super_1_en_4.3.0_3.0_1674212339330.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_super_1_en_4.3.0_3.0_1674212339330.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_super_1","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_super_1","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_super_1| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.1 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/nbroad/rob-base-superqa1 +- https://paperswithcode.com/sota?task=Question+Answering&dataset=adversarial_qa \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_super_2_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_super_2_en.md new file mode 100644 index 00000000000000..b573d0ea37758b --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_super_2_en.md @@ -0,0 +1,89 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from nbroad) +author: John Snow Labs +name: roberta_qa_base_super_2 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `rob-base-superqa2` is a English model originally trained by `nbroad`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_super_2_en_4.3.0_3.0_1674212395438.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_super_2_en_4.3.0_3.0_1674212395438.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_super_2","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_super_2","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_super_2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.1 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/nbroad/rob-base-superqa2 +- https://paperswithcode.com/sota?task=Question+Answering&dataset=squad_v2 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_task_specific_distilation_on_squad_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_task_specific_distilation_on_squad_en.md new file mode 100644 index 00000000000000..f6ff4b920e5f26 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_task_specific_distilation_on_squad_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_base_task_specific_distilation_on_squad +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `distilroberta-base-task-specific-distilation-on-squad` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_task_specific_distilation_on_squad_en_4.3.0_3.0_1674210519978.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_task_specific_distilation_on_squad_en_4.3.0_3.0_1674210519978.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_task_specific_distilation_on_squad","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_task_specific_distilation_on_squad","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_task_specific_distilation_on_squad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|307.1 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/distilroberta-base-task-specific-distilation-on-squad \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_tweet_model_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_tweet_model_en.md new file mode 100644 index 00000000000000..4ce45184023538 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_tweet_model_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from saburbutt) +author: John Snow Labs +name: roberta_qa_base_tweet_model +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta_base_tweetqa_model` is a English model originally trained by `saburbutt`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_tweet_model_en_4.3.0_3.0_1674223144072.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_tweet_model_en_4.3.0_3.0_1674223144072.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_tweet_model","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_tweet_model","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_tweet_model| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|432.7 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/saburbutt/roberta_base_tweetqa_model \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_uncased_squad_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_uncased_squad_en.md new file mode 100644 index 00000000000000..862892b86a046b --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_base_uncased_squad_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Uncased model (from t15) +author: John Snow Labs +name: roberta_qa_base_uncased_squad +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `distilroberta-base-uncased-squad` is a English model originally trained by `t15`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_base_uncased_squad_en_4.3.0_3.0_1674210559838.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_base_uncased_squad_en_4.3.0_3.0_1674210559838.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_uncased_squad","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_base_uncased_squad","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_base_uncased_squad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|307.1 MB| +|Case sensitive:|false| +|Max sentence length:|256| + +## References + +- https://huggingface.co/t15/distilroberta-base-uncased-squad \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_becasv4.1_es.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_becasv4.1_es.md new file mode 100644 index 00000000000000..9205d78fc7ee18 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_becasv4.1_es.md @@ -0,0 +1,88 @@ +--- +layout: model +title: Spanish RobertaForQuestionAnswering Cased model (from Evelyn18) +author: John Snow Labs +name: roberta_qa_becasv4.1 +date: 2023-01-20 +tags: [es, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: es +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `BECASV4.1` is a Spanish model originally trained by `Evelyn18`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_becasv4.1_es_4.3.0_3.0_1674207847912.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_becasv4.1_es_4.3.0_3.0_1674207847912.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_becasv4.1","es")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_becasv4.1","es") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_becasv4.1| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|es| +|Size:|459.9 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Evelyn18/BECASV4.1 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_bertserini_base_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_bertserini_base_en.md new file mode 100644 index 00000000000000..6f21982ef19df4 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_bertserini_base_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from rsvp-ai) +author: John Snow Labs +name: roberta_qa_bertserini_base +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `bertserini-roberta-base` is a English model originally trained by `rsvp-ai`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_bertserini_base_en_4.3.0_3.0_1674209264464.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_bertserini_base_en_4.3.0_3.0_1674209264464.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_bertserini_base","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_bertserini_base","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_bertserini_base| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|449.3 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/rsvp-ai/bertserini-roberta-base \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_ca_v2_squac_ca_catalan_ca.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_ca_v2_squac_ca_catalan_ca.md new file mode 100644 index 00000000000000..05e0e778e077bc --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_ca_v2_squac_ca_catalan_ca.md @@ -0,0 +1,88 @@ +--- +layout: model +title: Catalan RobertaForQuestionAnswering Cased model (from crodri) +author: John Snow Labs +name: roberta_qa_ca_v2_squac_ca_catalan +date: 2023-01-20 +tags: [ca, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: ca +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-ca-v2-qa-squac-ca-catalanqa` is a Catalan model originally trained by `crodri`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_ca_v2_squac_ca_catalan_ca_4.3.0_3.0_1674219909583.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_ca_v2_squac_ca_catalan_ca_4.3.0_3.0_1674219909583.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_ca_v2_squac_ca_catalan","ca")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_ca_v2_squac_ca_catalan","ca") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_ca_v2_squac_ca_catalan| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|ca| +|Size:|461.2 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/crodri/roberta-ca-v2-qa-squac-ca-catalanqa \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_canard_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_canard_en.md new file mode 100644 index 00000000000000..5dbb2e78b76ac2 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_canard_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from peggyhuang) +author: John Snow Labs +name: roberta_qa_canard +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-canard` is a English model originally trained by `peggyhuang`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_canard_en_4.3.0_3.0_1674219967411.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_canard_en_4.3.0_3.0_1674219967411.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_canard","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_canard","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_canard| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|465.1 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/peggyhuang/roberta-canard \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_checkpoint_finetuned_squad_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_checkpoint_finetuned_squad_en.md new file mode 100644 index 00000000000000..edaae508b23fdf --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_checkpoint_finetuned_squad_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from xinranyyyy) +author: John Snow Labs +name: roberta_qa_checkpoint_finetuned_squad +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta_checkpoint-finetuned-squad` is a English model originally trained by `xinranyyyy`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_checkpoint_finetuned_squad_en_4.3.0_3.0_1674223200816.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_checkpoint_finetuned_squad_en_4.3.0_3.0_1674223200816.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_checkpoint_finetuned_squad","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_checkpoint_finetuned_squad","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_checkpoint_finetuned_squad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|465.5 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/xinranyyyy/roberta_checkpoint-finetuned-squad \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_chiendvhust_base_finetuned_squad_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_chiendvhust_base_finetuned_squad_en.md new file mode 100644 index 00000000000000..dbd376bf09489f --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_chiendvhust_base_finetuned_squad_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from chiendvhust) +author: John Snow Labs +name: roberta_qa_chiendvhust_base_finetuned_squad +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-finetuned-squad` is a English model originally trained by `chiendvhust`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_chiendvhust_base_finetuned_squad_en_4.3.0_3.0_1674217182008.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_chiendvhust_base_finetuned_squad_en_4.3.0_3.0_1674217182008.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_chiendvhust_base_finetuned_squad","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_chiendvhust_base_finetuned_squad","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_chiendvhust_base_finetuned_squad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|457.5 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/chiendvhust/roberta-base-finetuned-squad \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_cline_emanuals_tech_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_cline_emanuals_tech_en.md new file mode 100644 index 00000000000000..4aa81b68b0003d --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_cline_emanuals_tech_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_cline_emanuals_tech +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `cline-emanuals-techqa` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_cline_emanuals_tech_en_4.3.0_3.0_1674209326690.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_cline_emanuals_tech_en_4.3.0_3.0_1674209326690.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_cline_emanuals_tech","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_cline_emanuals_tech","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_cline_emanuals_tech| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|466.4 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/cline-emanuals-techqa \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_custom_squad_ds_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_custom_squad_ds_en.md new file mode 100644 index 00000000000000..4f2ae8246caf05 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_custom_squad_ds_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from sunitha) +author: John Snow Labs +name: roberta_qa_custom_squad_ds +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `Roberta_Custom_Squad_DS` is a English model originally trained by `sunitha`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_custom_squad_ds_en_4.3.0_3.0_1674208785165.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_custom_squad_ds_en_4.3.0_3.0_1674208785165.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_custom_squad_ds","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_custom_squad_ds","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_custom_squad_ds| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.4 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/sunitha/Roberta_Custom_Squad_DS \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_customds_finetune_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_customds_finetune_en.md new file mode 100644 index 00000000000000..7eb207028a6d34 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_customds_finetune_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from sunitha) +author: John Snow Labs +name: roberta_qa_customds_finetune +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-customds-finetune` is a English model originally trained by `sunitha`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_customds_finetune_en_4.3.0_3.0_1674220035505.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_customds_finetune_en_4.3.0_3.0_1674220035505.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_customds_finetune","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_customds_finetune","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_customds_finetune| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.3 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/sunitha/roberta-customds-finetune \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_cv_custom_ds_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_cv_custom_ds_en.md new file mode 100644 index 00000000000000..9a12fbe60e2d95 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_cv_custom_ds_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from sunitha) +author: John Snow Labs +name: roberta_qa_cv_custom_ds +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `CV_Custom_DS` is a English model originally trained by `sunitha`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_cv_custom_ds_en_4.3.0_3.0_1674207905368.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_cv_custom_ds_en_4.3.0_3.0_1674207905368.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_cv_custom_ds","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_cv_custom_ds","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_cv_custom_ds| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.3 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/sunitha/CV_Custom_DS \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_cv_merge_ds_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_cv_merge_ds_en.md new file mode 100644 index 00000000000000..a3f2746b6b2cae --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_cv_merge_ds_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from sunitha) +author: John Snow Labs +name: roberta_qa_cv_merge_ds +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `CV_Merge_DS` is a English model originally trained by `sunitha`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_cv_merge_ds_en_4.3.0_3.0_1674207962257.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_cv_merge_ds_en_4.3.0_3.0_1674207962257.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_cv_merge_ds","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_cv_merge_ds","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_cv_merge_ds| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.3 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/sunitha/CV_Merge_DS \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_cyberlandr_door_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_cyberlandr_door_en.md new file mode 100644 index 00000000000000..4f40fdfa7b8962 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_cyberlandr_door_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from billfrench) +author: John Snow Labs +name: roberta_qa_cyberlandr_door +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `cyberlandr-door` is a English model originally trained by `billfrench`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_cyberlandr_door_en_4.3.0_3.0_1674209416902.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_cyberlandr_door_en_4.3.0_3.0_1674209416902.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_cyberlandr_door","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_cyberlandr_door","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_cyberlandr_door| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|414.0 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/billfrench/cyberlandr-door \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_deepakvk_base_squad2_finetuned_squad_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_deepakvk_base_squad2_finetuned_squad_en.md new file mode 100644 index 00000000000000..8d3b8b49fb5381 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_deepakvk_base_squad2_finetuned_squad_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from deepakvk) +author: John Snow Labs +name: roberta_qa_deepakvk_base_squad2_finetuned_squad +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-squad2-finetuned-squad` is a English model originally trained by `deepakvk`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_deepakvk_base_squad2_finetuned_squad_en_4.3.0_3.0_1674219250943.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_deepakvk_base_squad2_finetuned_squad_en_4.3.0_3.0_1674219250943.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_deepakvk_base_squad2_finetuned_squad","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_deepakvk_base_squad2_finetuned_squad","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_deepakvk_base_squad2_finetuned_squad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.3 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/deepakvk/roberta-base-squad2-finetuned-squad \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_deepset_base_squad2_orkg_how_1e_4_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_deepset_base_squad2_orkg_how_1e_4_en.md new file mode 100644 index 00000000000000..4bc782861e1ae1 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_deepset_base_squad2_orkg_how_1e_4_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from Moussab) +author: John Snow Labs +name: roberta_qa_deepset_base_squad2_orkg_how_1e_4 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `deepset-roberta-base-squad2-orkg-how-1e-4` is a English model originally trained by `Moussab`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_deepset_base_squad2_orkg_how_1e_4_en_4.3.0_3.0_1674209475140.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_deepset_base_squad2_orkg_how_1e_4_en_4.3.0_3.0_1674209475140.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_deepset_base_squad2_orkg_how_1e_4","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_deepset_base_squad2_orkg_how_1e_4","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_deepset_base_squad2_orkg_how_1e_4| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.8 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Moussab/deepset-roberta-base-squad2-orkg-how-1e-4 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_deepset_base_squad2_orkg_how_5e_05_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_deepset_base_squad2_orkg_how_5e_05_en.md new file mode 100644 index 00000000000000..4c41a1875c89de --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_deepset_base_squad2_orkg_how_5e_05_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from Moussab) +author: John Snow Labs +name: roberta_qa_deepset_base_squad2_orkg_how_5e_05 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `deepset-roberta-base-squad2-orkg-how-5e-05` is a English model originally trained by `Moussab`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_deepset_base_squad2_orkg_how_5e_05_en_4.3.0_3.0_1674209532316.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_deepset_base_squad2_orkg_how_5e_05_en_4.3.0_3.0_1674209532316.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_deepset_base_squad2_orkg_how_5e_05","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_deepset_base_squad2_orkg_how_5e_05","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_deepset_base_squad2_orkg_how_5e_05| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.8 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Moussab/deepset-roberta-base-squad2-orkg-how-5e-05 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_deepset_base_squad2_orkg_no_label_1e_4_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_deepset_base_squad2_orkg_no_label_1e_4_en.md new file mode 100644 index 00000000000000..caaccb428df110 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_deepset_base_squad2_orkg_no_label_1e_4_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from Moussab) +author: John Snow Labs +name: roberta_qa_deepset_base_squad2_orkg_no_label_1e_4 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `deepset-roberta-base-squad2-orkg-no-label-1e-4` is a English model originally trained by `Moussab`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_deepset_base_squad2_orkg_no_label_1e_4_en_4.3.0_3.0_1674209605485.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_deepset_base_squad2_orkg_no_label_1e_4_en_4.3.0_3.0_1674209605485.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_deepset_base_squad2_orkg_no_label_1e_4","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_deepset_base_squad2_orkg_no_label_1e_4","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_deepset_base_squad2_orkg_no_label_1e_4| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.9 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Moussab/deepset-roberta-base-squad2-orkg-no-label-1e-4 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_deepset_base_squad2_orkg_no_label_5e_05_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_deepset_base_squad2_orkg_no_label_5e_05_en.md new file mode 100644 index 00000000000000..938352b1d97b33 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_deepset_base_squad2_orkg_no_label_5e_05_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from Moussab) +author: John Snow Labs +name: roberta_qa_deepset_base_squad2_orkg_no_label_5e_05 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `deepset-roberta-base-squad2-orkg-no-label-5e-05` is a English model originally trained by `Moussab`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_deepset_base_squad2_orkg_no_label_5e_05_en_4.3.0_3.0_1674209664953.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_deepset_base_squad2_orkg_no_label_5e_05_en_4.3.0_3.0_1674209664953.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_deepset_base_squad2_orkg_no_label_5e_05","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_deepset_base_squad2_orkg_no_label_5e_05","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_deepset_base_squad2_orkg_no_label_5e_05| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.8 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Moussab/deepset-roberta-base-squad2-orkg-no-label-5e-05 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_deepset_base_squad2_orkg_what_1e_4_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_deepset_base_squad2_orkg_what_1e_4_en.md new file mode 100644 index 00000000000000..37f52ee2a4b38f --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_deepset_base_squad2_orkg_what_1e_4_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from Moussab) +author: John Snow Labs +name: roberta_qa_deepset_base_squad2_orkg_what_1e_4 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `deepset-roberta-base-squad2-orkg-what-1e-4` is a English model originally trained by `Moussab`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_deepset_base_squad2_orkg_what_1e_4_en_4.3.0_3.0_1674209722792.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_deepset_base_squad2_orkg_what_1e_4_en_4.3.0_3.0_1674209722792.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_deepset_base_squad2_orkg_what_1e_4","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_deepset_base_squad2_orkg_what_1e_4","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_deepset_base_squad2_orkg_what_1e_4| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.8 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Moussab/deepset-roberta-base-squad2-orkg-what-1e-4 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_deepset_base_squad2_orkg_which_1e_4_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_deepset_base_squad2_orkg_which_1e_4_en.md new file mode 100644 index 00000000000000..4f3ca4c5e441a1 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_deepset_base_squad2_orkg_which_1e_4_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from Moussab) +author: John Snow Labs +name: roberta_qa_deepset_base_squad2_orkg_which_1e_4 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `deepset-roberta-base-squad2-orkg-which-1e-4` is a English model originally trained by `Moussab`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_deepset_base_squad2_orkg_which_1e_4_en_4.3.0_3.0_1674209779299.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_deepset_base_squad2_orkg_which_1e_4_en_4.3.0_3.0_1674209779299.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_deepset_base_squad2_orkg_which_1e_4","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_deepset_base_squad2_orkg_which_1e_4","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_deepset_base_squad2_orkg_which_1e_4| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.8 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Moussab/deepset-roberta-base-squad2-orkg-which-1e-4 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_deepset_base_squad2_orkg_which_5e_05_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_deepset_base_squad2_orkg_which_5e_05_en.md new file mode 100644 index 00000000000000..6b777da35df3ce --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_deepset_base_squad2_orkg_which_5e_05_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from Moussab) +author: John Snow Labs +name: roberta_qa_deepset_base_squad2_orkg_which_5e_05 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `deepset-roberta-base-squad2-orkg-which-5e-05` is a English model originally trained by `Moussab`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_deepset_base_squad2_orkg_which_5e_05_en_4.3.0_3.0_1674209836673.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_deepset_base_squad2_orkg_which_5e_05_en_4.3.0_3.0_1674209836673.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_deepset_base_squad2_orkg_which_5e_05","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_deepset_base_squad2_orkg_which_5e_05","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_deepset_base_squad2_orkg_which_5e_05| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.8 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Moussab/deepset-roberta-base-squad2-orkg-which-5e-05 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_discord_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_discord_en.md new file mode 100644 index 00000000000000..6973bc1077ed86 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_discord_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from Salesforce) +author: John Snow Labs +name: roberta_qa_discord +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `discord_qa` is a English model originally trained by `Salesforce`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_discord_en_4.3.0_3.0_1674210253072.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_discord_en_4.3.0_3.0_1674210253072.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_discord","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_discord","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_discord| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|845.1 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Salesforce/discord_qa \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_distilrobert_base_squadv2_328seq_128stride_test_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_distilrobert_base_squadv2_328seq_128stride_test_en.md new file mode 100644 index 00000000000000..ab8bdafb9b025f --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_distilrobert_base_squadv2_328seq_128stride_test_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from manishiitg) +author: John Snow Labs +name: roberta_qa_distilrobert_base_squadv2_328seq_128stride_test +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `distilrobert-base-squadv2-328seq-128stride-test` is a English model originally trained by `manishiitg`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_distilrobert_base_squadv2_328seq_128stride_test_en_4.3.0_3.0_1674210316686.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_distilrobert_base_squadv2_328seq_128stride_test_en_4.3.0_3.0_1674210316686.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_distilrobert_base_squadv2_328seq_128stride_test","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_distilrobert_base_squadv2_328seq_128stride_test","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_distilrobert_base_squadv2_328seq_128stride_test| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|307.0 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/manishiitg/distilrobert-base-squadv2-328seq-128stride-test \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_dpr_nq_reader_base_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_dpr_nq_reader_base_en.md new file mode 100644 index 00000000000000..57db76612331a9 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_dpr_nq_reader_base_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from nlpconnect) +author: John Snow Labs +name: roberta_qa_dpr_nq_reader_base +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `dpr-nq-reader-roberta-base` is a English model originally trained by `nlpconnect`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_dpr_nq_reader_base_en_4.3.0_3.0_1674210699630.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_dpr_nq_reader_base_en_4.3.0_3.0_1674210699630.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_dpr_nq_reader_base","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_dpr_nq_reader_base","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_dpr_nq_reader_base| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|466.4 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/nlpconnect/dpr-nq-reader-roberta-base \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_dpr_nq_reader_base_v2_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_dpr_nq_reader_base_v2_en.md new file mode 100644 index 00000000000000..4e5671d1d0c898 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_dpr_nq_reader_base_v2_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from nlpconnect) +author: John Snow Labs +name: roberta_qa_dpr_nq_reader_base_v2 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `dpr-nq-reader-roberta-base-v2` is a English model originally trained by `nlpconnect`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_dpr_nq_reader_base_v2_en_4.3.0_3.0_1674210757247.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_dpr_nq_reader_base_v2_en_4.3.0_3.0_1674210757247.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_dpr_nq_reader_base_v2","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_dpr_nq_reader_base_v2","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_dpr_nq_reader_base_v2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|466.3 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/nlpconnect/dpr-nq-reader-roberta-base-v2 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_dynamic_minilmv2_l6_h384_squad1.1_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_dynamic_minilmv2_l6_h384_squad1.1_en.md new file mode 100644 index 00000000000000..8dcfa7b4418c19 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_dynamic_minilmv2_l6_h384_squad1.1_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Mini Cased model (from sguskin) +author: John Snow Labs +name: roberta_qa_dynamic_minilmv2_l6_h384_squad1.1 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `dynamic-minilmv2-L6-H384-squad1.1` is a English model originally trained by `sguskin`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_dynamic_minilmv2_l6_h384_squad1.1_en_4.3.0_3.0_1674210790213.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_dynamic_minilmv2_l6_h384_squad1.1_en_4.3.0_3.0_1674210790213.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_dynamic_minilmv2_l6_h384_squad1.1","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_dynamic_minilmv2_l6_h384_squad1.1","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_dynamic_minilmv2_l6_h384_squad1.1| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|112.1 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/sguskin/dynamic-minilmv2-L6-H384-squad1.1 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_eda_and_parav3_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_eda_and_parav3_en.md new file mode 100644 index 00000000000000..1436f1dcdff391 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_eda_and_parav3_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from comacrae) +author: John Snow Labs +name: roberta_qa_eda_and_parav3 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-eda-and-parav3` is a English model originally trained by `comacrae`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_eda_and_parav3_en_4.3.0_3.0_1674220091398.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_eda_and_parav3_en_4.3.0_3.0_1674220091398.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_eda_and_parav3","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_eda_and_parav3","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_eda_and_parav3| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.5 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/comacrae/roberta-eda-and-parav3 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_edav3_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_edav3_en.md new file mode 100644 index 00000000000000..8ff9a7a6058449 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_edav3_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from comacrae) +author: John Snow Labs +name: roberta_qa_edav3 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-edav3` is a English model originally trained by `comacrae`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_edav3_en_4.3.0_3.0_1674220150986.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_edav3_en_4.3.0_3.0_1674220150986.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_edav3","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_edav3","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_edav3| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.5 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/comacrae/roberta-edav3 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_edtech_model1_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_edtech_model1_en.md new file mode 100644 index 00000000000000..5d04d4e4e00711 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_edtech_model1_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from phanimvsk) +author: John Snow Labs +name: roberta_qa_edtech_model1 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `Edtech_model1` is a English model originally trained by `phanimvsk`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_edtech_model1_en_4.3.0_3.0_1674208077877.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_edtech_model1_en_4.3.0_3.0_1674208077877.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_edtech_model1","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_edtech_model1","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_edtech_model1| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|463.3 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/phanimvsk/Edtech_model1 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_emanuals_squad2.0_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_emanuals_squad2.0_en.md new file mode 100644 index 00000000000000..8abe6b55980b23 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_emanuals_squad2.0_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_emanuals_squad2.0 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `EManuals_RoBERTa_squad2.0` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_emanuals_squad2.0_en_4.3.0_3.0_1674208019986.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_emanuals_squad2.0_en_4.3.0_3.0_1674208019986.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_emanuals_squad2.0","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_emanuals_squad2.0","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_emanuals_squad2.0| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|466.7 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/EManuals_RoBERTa_squad2.0 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_emotion_extraction_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_emotion_extraction_en.md new file mode 100644 index 00000000000000..ed449c2ea133d4 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_emotion_extraction_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from Nakul24) +author: John Snow Labs +name: roberta_qa_emotion_extraction +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `RoBERTa-emotion-extraction` is a English model originally trained by `Nakul24`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_emotion_extraction_en_4.3.0_3.0_1674208609562.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_emotion_extraction_en_4.3.0_3.0_1674208609562.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_emotion_extraction","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_emotion_extraction","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_emotion_extraction| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|426.9 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Nakul24/RoBERTa-emotion-extraction \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_fin_v1_finetuned_squad_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_fin_v1_finetuned_squad_en.md new file mode 100644 index 00000000000000..16397b6e8f0156 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_fin_v1_finetuned_squad_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from cgou) +author: John Snow Labs +name: roberta_qa_fin_v1_finetuned_squad +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `fin_RoBERTa-v1-finetuned-squad` is a English model originally trained by `cgou`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_fin_v1_finetuned_squad_en_4.3.0_3.0_1674210819999.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_fin_v1_finetuned_squad_en_4.3.0_3.0_1674210819999.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_fin_v1_finetuned_squad","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_fin_v1_finetuned_squad","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_fin_v1_finetuned_squad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|248.1 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/cgou/fin_RoBERTa-v1-finetuned-squad \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_finetuned_city_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_finetuned_city_en.md new file mode 100644 index 00000000000000..69478a5f05b47a --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_finetuned_city_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from skandaonsolve) +author: John Snow Labs +name: roberta_qa_finetuned_city +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-finetuned-city` is a English model originally trained by `skandaonsolve`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_finetuned_city_en_4.3.0_3.0_1674220204421.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_finetuned_city_en_4.3.0_3.0_1674220204421.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_finetuned_city","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_finetuned_city","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_finetuned_city| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.7 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/skandaonsolve/roberta-finetuned-city \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_finetuned_country_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_finetuned_country_en.md new file mode 100644 index 00000000000000..513f06bb743ff1 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_finetuned_country_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from nsridhar) +author: John Snow Labs +name: roberta_qa_finetuned_country +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-finetuned-country` is a English model originally trained by `nsridhar`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_finetuned_country_en_4.3.0_3.0_1674220262882.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_finetuned_country_en_4.3.0_3.0_1674220262882.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_finetuned_country","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_finetuned_country","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_finetuned_country| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.4 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/nsridhar/roberta-finetuned-country \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_finetuned_facility_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_finetuned_facility_en.md new file mode 100644 index 00000000000000..b9a23353e92cdc --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_finetuned_facility_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from skandaonsolve) +author: John Snow Labs +name: roberta_qa_finetuned_facility +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-finetuned-facility` is a English model originally trained by `skandaonsolve`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_finetuned_facility_en_4.3.0_3.0_1674220319873.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_finetuned_facility_en_4.3.0_3.0_1674220319873.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_finetuned_facility","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_finetuned_facility","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_finetuned_facility| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.4 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/skandaonsolve/roberta-finetuned-facility \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_finetuned_location_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_finetuned_location_en.md new file mode 100644 index 00000000000000..9c963c43511b4d --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_finetuned_location_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from skandaonsolve) +author: John Snow Labs +name: roberta_qa_finetuned_location +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-finetuned-location` is a English model originally trained by `skandaonsolve`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_finetuned_location_en_4.3.0_3.0_1674220382399.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_finetuned_location_en_4.3.0_3.0_1674220382399.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_finetuned_location","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_finetuned_location","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_finetuned_location| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.7 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/skandaonsolve/roberta-finetuned-location \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_finetuned_squad_50k_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_finetuned_squad_50k_en.md new file mode 100644 index 00000000000000..9abc5fbefcbd53 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_finetuned_squad_50k_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from clementgyj) +author: John Snow Labs +name: roberta_qa_finetuned_squad_50k +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-finetuned-squad-50k` is a English model originally trained by `clementgyj`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_finetuned_squad_50k_en_4.3.0_3.0_1674220438911.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_finetuned_squad_50k_en_4.3.0_3.0_1674220438911.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_finetuned_squad_50k","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_finetuned_squad_50k","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_finetuned_squad_50k| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|462.4 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/clementgyj/roberta-finetuned-squad-50k \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_finetuned_squadv1_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_finetuned_squadv1_en.md new file mode 100644 index 00000000000000..6ad9e19eaa2351 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_finetuned_squadv1_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from mrm8488) +author: John Snow Labs +name: roberta_qa_finetuned_squadv1 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `distilroberta-finetuned-squadv1` is a English model originally trained by `mrm8488`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_finetuned_squadv1_en_4.3.0_3.0_1674210647091.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_finetuned_squadv1_en_4.3.0_3.0_1674210647091.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_finetuned_squadv1","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_finetuned_squadv1","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_finetuned_squadv1| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|307.1 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/mrm8488/distilroberta-finetuned-squadv1 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_finetuned_state2_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_finetuned_state2_en.md new file mode 100644 index 00000000000000..9e0c551ec6c5d5 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_finetuned_state2_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from skandaonsolve) +author: John Snow Labs +name: roberta_qa_finetuned_state2 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-finetuned-state2` is a English model originally trained by `skandaonsolve`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_finetuned_state2_en_4.3.0_3.0_1674220557287.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_finetuned_state2_en_4.3.0_3.0_1674220557287.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_finetuned_state2","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_finetuned_state2","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_finetuned_state2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.5 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/skandaonsolve/roberta-finetuned-state2 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_finetuned_state_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_finetuned_state_en.md new file mode 100644 index 00000000000000..b7ab7bf4ef94c0 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_finetuned_state_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from skandaonsolve) +author: John Snow Labs +name: roberta_qa_finetuned_state +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-finetuned-state` is a English model originally trained by `skandaonsolve`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_finetuned_state_en_4.3.0_3.0_1674220497335.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_finetuned_state_en_4.3.0_3.0_1674220497335.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_finetuned_state","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_finetuned_state","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_finetuned_state| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.5 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/skandaonsolve/roberta-finetuned-state \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_finetuned_timeentities2_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_finetuned_timeentities2_en.md new file mode 100644 index 00000000000000..2ecd5e4d0a0229 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_finetuned_timeentities2_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from skandaonsolve) +author: John Snow Labs +name: roberta_qa_finetuned_timeentities2 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-finetuned-timeentities2` is a English model originally trained by `skandaonsolve`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_finetuned_timeentities2_en_4.3.0_3.0_1674220671794.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_finetuned_timeentities2_en_4.3.0_3.0_1674220671794.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_finetuned_timeentities2","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_finetuned_timeentities2","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_finetuned_timeentities2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|465.0 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/skandaonsolve/roberta-finetuned-timeentities2 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_finetuned_timeentities2_ttsp75_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_finetuned_timeentities2_ttsp75_en.md new file mode 100644 index 00000000000000..619b29c0205e2f --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_finetuned_timeentities2_ttsp75_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from skandaonsolve) +author: John Snow Labs +name: roberta_qa_finetuned_timeentities2_ttsp75 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-finetuned-timeentities2_ttsp75` is a English model originally trained by `skandaonsolve`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_finetuned_timeentities2_ttsp75_en_4.3.0_3.0_1674220728523.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_finetuned_timeentities2_ttsp75_en_4.3.0_3.0_1674220728523.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_finetuned_timeentities2_ttsp75","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_finetuned_timeentities2_ttsp75","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_finetuned_timeentities2_ttsp75| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.9 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/skandaonsolve/roberta-finetuned-timeentities2_ttsp75 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_finetuned_timeentities_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_finetuned_timeentities_en.md new file mode 100644 index 00000000000000..530d6a3f5f3d3f --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_finetuned_timeentities_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from skandaonsolve) +author: John Snow Labs +name: roberta_qa_finetuned_timeentities +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-finetuned-timeentities` is a English model originally trained by `skandaonsolve`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_finetuned_timeentities_en_4.3.0_3.0_1674220613032.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_finetuned_timeentities_en_4.3.0_3.0_1674220613032.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_finetuned_timeentities","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_finetuned_timeentities","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_finetuned_timeentities| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.9 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/skandaonsolve/roberta-finetuned-timeentities \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_firat_base_finetuned_squad_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_firat_base_finetuned_squad_en.md new file mode 100644 index 00000000000000..04621300db5a1c --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_firat_base_finetuned_squad_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from Firat) +author: John Snow Labs +name: roberta_qa_firat_base_finetuned_squad +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-finetuned-squad` is a English model originally trained by `Firat`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_firat_base_finetuned_squad_en_4.3.0_3.0_1674217059391.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_firat_base_finetuned_squad_en_4.3.0_3.0_1674217059391.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_firat_base_finetuned_squad","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_firat_base_finetuned_squad","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_firat_base_finetuned_squad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.2 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Firat/roberta-base-finetuned-squad \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_for_event_extraction_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_for_event_extraction_en.md new file mode 100644 index 00000000000000..125773a9cb8e93 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_for_event_extraction_en.md @@ -0,0 +1,91 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from veronica320) +author: John Snow Labs +name: roberta_qa_for_event_extraction +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `QA-for-Event-Extraction` is a English model originally trained by `veronica320`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_for_event_extraction_en_4.3.0_3.0_1674208285851.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_for_event_extraction_en_4.3.0_3.0_1674208285851.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_for_event_extraction","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_for_event_extraction","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_for_event_extraction| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|1.3 GB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/veronica320/QA-for-Event-Extraction +- https://aclanthology.org/2021.acl-short.42/ +- https://github.com/uwnlp/qamr +- https://github.com/veronica320/Zeroshot-Event-Extraction \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_for_question_answering_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_for_question_answering_en.md new file mode 100644 index 00000000000000..a602a63f314ab1 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_for_question_answering_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from Zamachi) +author: John Snow Labs +name: roberta_qa_for_question_answering +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-for-question-answering` is a English model originally trained by `Zamachi`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_for_question_answering_en_4.3.0_3.0_1674220787682.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_for_question_answering_en_4.3.0_3.0_1674220787682.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_for_question_answering","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_for_question_answering","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_for_question_answering| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|466.1 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Zamachi/roberta-for-question-answering \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_for_seizurefrequency_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_for_seizurefrequency_en.md new file mode 100644 index 00000000000000..d1fbf61e7dac33 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_for_seizurefrequency_en.md @@ -0,0 +1,89 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from CNT-UPenn) +author: John Snow Labs +name: roberta_qa_for_seizurefrequency +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `RoBERTa_for_seizureFrequency_QA` is a English model originally trained by `CNT-UPenn`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_for_seizurefrequency_en_4.3.0_3.0_1674208667059.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_for_seizurefrequency_en_4.3.0_3.0_1674208667059.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_for_seizurefrequency","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_for_seizurefrequency","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_for_seizurefrequency| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|466.9 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/CNT-UPenn/RoBERTa_for_seizureFrequency_QA +- https://doi.org/10.1093/jamia/ocac018 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_fpdm_ft_news_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_fpdm_ft_news_en.md new file mode 100644 index 00000000000000..4d00701cf521d1 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_fpdm_ft_news_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_fpdm_ft_news +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `fpdm_roberta_FT_newsqa` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_fpdm_ft_news_en_4.3.0_3.0_1674211000201.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_fpdm_ft_news_en_4.3.0_3.0_1674211000201.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_fpdm_ft_news","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_fpdm_ft_news","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_fpdm_ft_news| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|458.6 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/fpdm_roberta_FT_newsqa \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_fpdm_hier_ft_new_news_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_fpdm_hier_ft_new_news_en.md new file mode 100644 index 00000000000000..900c227c7e554d --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_fpdm_hier_ft_new_news_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_fpdm_hier_ft_new_news +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `fpdm_hier_roberta_FT_new_newsqa` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_fpdm_hier_ft_new_news_en_4.3.0_3.0_1674210874319.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_fpdm_hier_ft_new_news_en_4.3.0_3.0_1674210874319.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_fpdm_hier_ft_new_news","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_fpdm_hier_ft_new_news","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_fpdm_hier_ft_new_news| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|461.5 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/fpdm_hier_roberta_FT_new_newsqa \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_fpdm_hier_ft_news_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_fpdm_hier_ft_news_en.md new file mode 100644 index 00000000000000..e7abe85ac47931 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_fpdm_hier_ft_news_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_fpdm_hier_ft_news +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `fpdm_hier_roberta_FT_newsqa` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_fpdm_hier_ft_news_en_4.3.0_3.0_1674210937245.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_fpdm_hier_ft_news_en_4.3.0_3.0_1674210937245.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_fpdm_hier_ft_news","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_fpdm_hier_ft_news","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_fpdm_hier_ft_news| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|458.6 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/fpdm_hier_roberta_FT_newsqa \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_fpdm_pert_sent_0.01_squad2.0_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_fpdm_pert_sent_0.01_squad2.0_en.md new file mode 100644 index 00000000000000..352507f59e88dd --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_fpdm_pert_sent_0.01_squad2.0_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_fpdm_pert_sent_0.01_squad2.0 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `fpdm_roberta_pert_sent_0.01_squad2.0` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_fpdm_pert_sent_0.01_squad2.0_en_4.3.0_3.0_1674211060807.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_fpdm_pert_sent_0.01_squad2.0_en_4.3.0_3.0_1674211060807.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_fpdm_pert_sent_0.01_squad2.0","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_fpdm_pert_sent_0.01_squad2.0","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_fpdm_pert_sent_0.01_squad2.0| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|460.4 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/fpdm_roberta_pert_sent_0.01_squad2.0 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_fpdm_soup_model_squad2.0_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_fpdm_soup_model_squad2.0_en.md new file mode 100644 index 00000000000000..b1fcd37c30d2d3 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_fpdm_soup_model_squad2.0_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_fpdm_soup_model_squad2.0 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `fpdm_roberta_soup_model_squad2.0` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_fpdm_soup_model_squad2.0_en_4.3.0_3.0_1674211121560.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_fpdm_soup_model_squad2.0_en_4.3.0_3.0_1674211121560.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_fpdm_soup_model_squad2.0","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_fpdm_soup_model_squad2.0","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_fpdm_soup_model_squad2.0| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|460.4 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/fpdm_roberta_soup_model_squad2.0 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_fpdm_triplet_ft_new_news_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_fpdm_triplet_ft_new_news_en.md new file mode 100644 index 00000000000000..2a592cd156ed4d --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_fpdm_triplet_ft_new_news_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_fpdm_triplet_ft_new_news +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `fpdm_triplet_roberta_FT_new_newsqa` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_fpdm_triplet_ft_new_news_en_4.3.0_3.0_1674211184082.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_fpdm_triplet_ft_new_news_en_4.3.0_3.0_1674211184082.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_fpdm_triplet_ft_new_news","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_fpdm_triplet_ft_new_news","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_fpdm_triplet_ft_new_news| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|461.5 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/fpdm_triplet_roberta_FT_new_newsqa \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_fpdm_triplet_ft_news_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_fpdm_triplet_ft_news_en.md new file mode 100644 index 00000000000000..39c8596ec5d3b8 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_fpdm_triplet_ft_news_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_fpdm_triplet_ft_news +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `fpdm_triplet_roberta_FT_newsqa` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_fpdm_triplet_ft_news_en_4.3.0_3.0_1674211248579.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_fpdm_triplet_ft_news_en_4.3.0_3.0_1674211248579.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_fpdm_triplet_ft_news","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_fpdm_triplet_ft_news","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_fpdm_triplet_ft_news| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|458.6 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/fpdm_triplet_roberta_FT_newsqa \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_ft_new_news_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_ft_new_news_en.md new file mode 100644 index 00000000000000..f3939188e05659 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_ft_new_news_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_ft_new_news +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta_FT_new_newsqa` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_ft_new_news_en_4.3.0_3.0_1674222919626.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_ft_new_news_en_4.3.0_3.0_1674222919626.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_ft_new_news","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_ft_new_news","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_ft_new_news| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|461.5 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/roberta_FT_new_newsqa \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_ft_news_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_ft_news_en.md new file mode 100644 index 00000000000000..577c94b03a3e29 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_ft_news_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_ft_news +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta_FT_newsqa` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_ft_news_en_4.3.0_3.0_1674222981909.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_ft_news_en_4.3.0_3.0_1674222981909.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_ft_news","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_ft_news","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_ft_news| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|458.6 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/roberta_FT_newsqa \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_huxxx657_base_finetuned_squad_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_huxxx657_base_finetuned_squad_en.md new file mode 100644 index 00000000000000..0ee56332c583f1 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_huxxx657_base_finetuned_squad_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from huxxx657) +author: John Snow Labs +name: roberta_qa_huxxx657_base_finetuned_squad +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-finetuned-squad` is a English model originally trained by `huxxx657`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_huxxx657_base_finetuned_squad_en_4.3.0_3.0_1674217238547.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_huxxx657_base_finetuned_squad_en_4.3.0_3.0_1674217238547.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_huxxx657_base_finetuned_squad","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_huxxx657_base_finetuned_squad","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_huxxx657_base_finetuned_squad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.3 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/huxxx657/roberta-base-finetuned-squad \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_icebert_is_finetune_is.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_icebert_is_finetune_is.md new file mode 100644 index 00000000000000..802f84ff363ab3 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_icebert_is_finetune_is.md @@ -0,0 +1,88 @@ +--- +layout: model +title: Icelandic RobertaForQuestionAnswering Cased model (from nozagleh) +author: John Snow Labs +name: roberta_qa_icebert_is_finetune +date: 2023-01-20 +tags: [is, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: is +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `IceBERT-QA-Is-finetune` is a Icelandic model originally trained by `nozagleh`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_icebert_is_finetune_is_4.3.0_3.0_1674208136981.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_icebert_is_finetune_is_4.3.0_3.0_1674208136981.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_icebert_is_finetune","is")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_icebert_is_finetune","is") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_icebert_is_finetune| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|is| +|Size:|451.0 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/nozagleh/IceBERT-QA-Is-finetune \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_janeel_base_finetuned_squad_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_janeel_base_finetuned_squad_en.md new file mode 100644 index 00000000000000..c3bba4b0605adb --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_janeel_base_finetuned_squad_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from janeel) +author: John Snow Labs +name: roberta_qa_janeel_base_finetuned_squad +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-finetuned-squad` is a English model originally trained by `janeel`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_janeel_base_finetuned_squad_en_4.3.0_3.0_1674217296605.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_janeel_base_finetuned_squad_en_4.3.0_3.0_1674217296605.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_janeel_base_finetuned_squad","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_janeel_base_finetuned_squad","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_janeel_base_finetuned_squad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.3 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/janeel/roberta-base-finetuned-squad \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_janeel_tiny_squad2_finetuned_squad_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_janeel_tiny_squad2_finetuned_squad_en.md new file mode 100644 index 00000000000000..dac2092286b6b3 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_janeel_tiny_squad2_finetuned_squad_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Tiny Cased model (from janeel) +author: John Snow Labs +name: roberta_qa_janeel_tiny_squad2_finetuned_squad +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `tinyroberta-squad2-finetuned-squad` is a English model originally trained by `janeel`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_janeel_tiny_squad2_finetuned_squad_en_4.3.0_3.0_1674224399345.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_janeel_tiny_squad2_finetuned_squad_en_4.3.0_3.0_1674224399345.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_janeel_tiny_squad2_finetuned_squad","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_janeel_tiny_squad2_finetuned_squad","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_janeel_tiny_squad2_finetuned_squad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|307.3 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/janeel/tinyroberta-squad2-finetuned-squad \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_jgammack_base_squad_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_jgammack_base_squad_en.md new file mode 100644 index 00000000000000..1bbce8ede82dd3 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_jgammack_base_squad_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from jgammack) +author: John Snow Labs +name: roberta_qa_jgammack_base_squad +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-squad` is a English model originally trained by `jgammack`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_jgammack_base_squad_en_4.3.0_3.0_1674218670079.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_jgammack_base_squad_en_4.3.0_3.0_1674218670079.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_jgammack_base_squad","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_jgammack_base_squad","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_jgammack_base_squad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.2 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/jgammack/roberta-base-squad \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_joantirant_base_bne_finetuned_s_c_es.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_joantirant_base_bne_finetuned_s_c_es.md new file mode 100644 index 00000000000000..ae2e3ee848740f --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_joantirant_base_bne_finetuned_s_c_es.md @@ -0,0 +1,88 @@ +--- +layout: model +title: Spanish RobertaForQuestionAnswering Base Cased model (from JoanTirant) +author: John Snow Labs +name: roberta_qa_joantirant_base_bne_finetuned_s_c +date: 2023-01-20 +tags: [es, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: es +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-bne-finetuned-sqac` is a Spanish model originally trained by `JoanTirant`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_joantirant_base_bne_finetuned_s_c_es_4.3.0_3.0_1674212952508.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_joantirant_base_bne_finetuned_s_c_es_4.3.0_3.0_1674212952508.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_joantirant_base_bne_finetuned_s_c","es")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_joantirant_base_bne_finetuned_s_c","es") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_joantirant_base_bne_finetuned_s_c| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|es| +|Size:|460.4 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/JoanTirant/roberta-base-bne-finetuned-sqac \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_jonatangk_base_bne_finetuned_s_c_es.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_jonatangk_base_bne_finetuned_s_c_es.md new file mode 100644 index 00000000000000..20e792e2ada72e --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_jonatangk_base_bne_finetuned_s_c_es.md @@ -0,0 +1,88 @@ +--- +layout: model +title: Spanish RobertaForQuestionAnswering Base Cased model (from JonatanGk) +author: John Snow Labs +name: roberta_qa_jonatangk_base_bne_finetuned_s_c +date: 2023-01-20 +tags: [es, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: es +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-bne-finetuned-sqac` is a Spanish model originally trained by `JonatanGk`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_jonatangk_base_bne_finetuned_s_c_es_4.3.0_3.0_1674213010026.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_jonatangk_base_bne_finetuned_s_c_es_4.3.0_3.0_1674213010026.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_jonatangk_base_bne_finetuned_s_c","es")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_jonatangk_base_bne_finetuned_s_c","es") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_jonatangk_base_bne_finetuned_s_c| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|es| +|Size:|460.4 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/JonatanGk/roberta-base-bne-finetuned-sqac \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_l_squadv1.1_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_l_squadv1.1_en.md new file mode 100644 index 00000000000000..58a9b4828f7f0c --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_l_squadv1.1_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from vuiseng9) +author: John Snow Labs +name: roberta_qa_l_squadv1.1 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-l-squadv1.1` is a English model originally trained by `vuiseng9`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_l_squadv1.1_en_4.3.0_3.0_1674220913524.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_l_squadv1.1_en_4.3.0_3.0_1674220913524.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_l_squadv1.1","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_l_squadv1.1","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_l_squadv1.1| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|1.3 GB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/vuiseng9/roberta-l-squadv1.1 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_large_data_seed_0_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_large_data_seed_0_en.md new file mode 100644 index 00000000000000..31fd24d215ab48 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_large_data_seed_0_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Large Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_large_data_seed_0 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-large-data-seed-0` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_large_data_seed_0_en_4.3.0_3.0_1674221077269.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_large_data_seed_0_en_4.3.0_3.0_1674221077269.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_large_data_seed_0","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_large_data_seed_0","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_large_data_seed_0| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|1.3 GB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-large-data-seed-0 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_large_data_seed_4_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_large_data_seed_4_en.md new file mode 100644 index 00000000000000..be6e760daa764b --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_large_data_seed_4_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Large Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_large_data_seed_4 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-large-data-seed-4` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_large_data_seed_4_en_4.3.0_3.0_1674221236388.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_large_data_seed_4_en_4.3.0_3.0_1674221236388.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_large_data_seed_4","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_large_data_seed_4","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_large_data_seed_4| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|1.3 GB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-large-data-seed-4 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_large_few_shot_k_1024_finetuned_squad_seed_4_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_large_few_shot_k_1024_finetuned_squad_seed_4_en.md new file mode 100644 index 00000000000000..acb9c127b4108d --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_large_few_shot_k_1024_finetuned_squad_seed_4_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Large Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_large_few_shot_k_1024_finetuned_squad_seed_4 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-large-few-shot-k-1024-finetuned-squad-seed-4` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_large_few_shot_k_1024_finetuned_squad_seed_4_en_4.3.0_3.0_1674221411536.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_large_few_shot_k_1024_finetuned_squad_seed_4_en_4.3.0_3.0_1674221411536.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_large_few_shot_k_1024_finetuned_squad_seed_4","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_large_few_shot_k_1024_finetuned_squad_seed_4","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_large_few_shot_k_1024_finetuned_squad_seed_4| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|1.3 GB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-large-few-shot-k-1024-finetuned-squad-seed-4 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_large_few_shot_k_32_finetuned_squad_seed_0_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_large_few_shot_k_32_finetuned_squad_seed_0_en.md new file mode 100644 index 00000000000000..0fbded5786ca73 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_large_few_shot_k_32_finetuned_squad_seed_0_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Large Cased model (from anas-awadalla) +author: John Snow Labs +name: roberta_qa_large_few_shot_k_32_finetuned_squad_seed_0 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-large-few-shot-k-32-finetuned-squad-seed-0` is a English model originally trained by `anas-awadalla`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_large_few_shot_k_32_finetuned_squad_seed_0_en_4.3.0_3.0_1674221604163.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_large_few_shot_k_32_finetuned_squad_seed_0_en_4.3.0_3.0_1674221604163.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_large_few_shot_k_32_finetuned_squad_seed_0","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_large_few_shot_k_32_finetuned_squad_seed_0","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_large_few_shot_k_32_finetuned_squad_seed_0| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|1.3 GB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anas-awadalla/roberta-large-few-shot-k-32-finetuned-squad-seed-0 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_large_fine_tuned_squad_es.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_large_fine_tuned_squad_es.md new file mode 100644 index 00000000000000..56b9617a5c99ae --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_large_fine_tuned_squad_es.md @@ -0,0 +1,88 @@ +--- +layout: model +title: Spanish RobertaForQuestionAnswering Large Cased model (from stevemobs) +author: John Snow Labs +name: roberta_qa_large_fine_tuned_squad +date: 2023-01-20 +tags: [es, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: es +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-large-fine-tuned-squad-es` is a Spanish model originally trained by `stevemobs`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_large_fine_tuned_squad_es_4.3.0_3.0_1674221753097.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_large_fine_tuned_squad_es_4.3.0_3.0_1674221753097.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_large_fine_tuned_squad","es")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_large_fine_tuned_squad","es") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_large_fine_tuned_squad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|es| +|Size:|1.3 GB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/stevemobs/roberta-large-fine-tuned-squad-es \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_large_squad2_hp_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_large_squad2_hp_en.md new file mode 100644 index 00000000000000..1f461e817ba1c1 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_large_squad2_hp_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Large Cased model (from deepset) +author: John Snow Labs +name: roberta_qa_large_squad2_hp +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-large-squad2-hp` is a English model originally trained by `deepset`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_large_squad2_hp_en_4.3.0_3.0_1674222219863.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_large_squad2_hp_en_4.3.0_3.0_1674222219863.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_large_squad2_hp","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_large_squad2_hp","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_large_squad2_hp| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|1.3 GB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/deepset/roberta-large-squad2-hp \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_large_squad_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_large_squad_en.md new file mode 100644 index 00000000000000..ab634182909f46 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_large_squad_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Large Cased model (from susghosh) +author: John Snow Labs +name: roberta_qa_large_squad +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-large-squad` is a English model originally trained by `susghosh`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_large_squad_en_4.3.0_3.0_1674221913718.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_large_squad_en_4.3.0_3.0_1674221913718.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_large_squad","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_large_squad","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_large_squad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|1.3 GB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/susghosh/roberta-large-squad \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_large_squad_v1_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_large_squad_v1_en.md new file mode 100644 index 00000000000000..e01c9ccce508a3 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_large_squad_v1_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Large Cased model (from csarron) +author: John Snow Labs +name: roberta_qa_large_squad_v1 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-large-squad-v1` is a English model originally trained by `csarron`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_large_squad_v1_en_4.3.0_3.0_1674222069233.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_large_squad_v1_en_4.3.0_3.0_1674222069233.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_large_squad_v1","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_large_squad_v1","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_large_squad_v1| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|1.3 GB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/csarron/roberta-large-squad-v1 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_li_base_squad2_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_li_base_squad2_en.md new file mode 100644 index 00000000000000..ee416ca7bae6d4 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_li_base_squad2_en.md @@ -0,0 +1,90 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from Li) +author: John Snow Labs +name: roberta_qa_li_base_squad2 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-squad2` is a English model originally trained by `Li`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_li_base_squad2_en_4.3.0_3.0_1674218901284.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_li_base_squad2_en_4.3.0_3.0_1674218901284.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_li_base_squad2","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_li_base_squad2","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_li_base_squad2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|462.0 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Li/roberta-base-squad2 +- https://rajpurkar.github.io/SQuAD-explorer +- https://rajpurkar.github.io/SQuAD-explorer/ \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_linh101201_base_finetuned_squad_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_linh101201_base_finetuned_squad_en.md new file mode 100644 index 00000000000000..97a58f858b5527 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_linh101201_base_finetuned_squad_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from linh101201) +author: John Snow Labs +name: roberta_qa_linh101201_base_finetuned_squad +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-finetuned-squad` is a English model originally trained by `linh101201`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_linh101201_base_finetuned_squad_en_4.3.0_3.0_1674217360099.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_linh101201_base_finetuned_squad_en_4.3.0_3.0_1674217360099.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_linh101201_base_finetuned_squad","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_linh101201_base_finetuned_squad","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_linh101201_base_finetuned_squad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|424.0 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/linh101201/roberta-base-finetuned-squad \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_lorenzkuhn_base_finetuned_squad_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_lorenzkuhn_base_finetuned_squad_en.md new file mode 100644 index 00000000000000..a56872c75b5b1a --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_lorenzkuhn_base_finetuned_squad_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from lorenzkuhn) +author: John Snow Labs +name: roberta_qa_lorenzkuhn_base_finetuned_squad +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-finetuned-squad` is a English model originally trained by `lorenzkuhn`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_lorenzkuhn_base_finetuned_squad_en_4.3.0_3.0_1674217419784.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_lorenzkuhn_base_finetuned_squad_en_4.3.0_3.0_1674217419784.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_lorenzkuhn_base_finetuned_squad","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_lorenzkuhn_base_finetuned_squad","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_lorenzkuhn_base_finetuned_squad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.2 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/lorenzkuhn/roberta-base-finetuned-squad \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_mask_step_pretraining_base_squadv2_epochs_3_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_mask_step_pretraining_base_squadv2_epochs_3_en.md new file mode 100644 index 00000000000000..8aefa83f5c69a4 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_mask_step_pretraining_base_squadv2_epochs_3_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_mask_step_pretraining_base_squadv2_epochs_3 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `mask_step_pretraining_roberta-base_squadv2_epochs_3` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_mask_step_pretraining_base_squadv2_epochs_3_en_4.3.0_3.0_1674211403400.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_mask_step_pretraining_base_squadv2_epochs_3_en_4.3.0_3.0_1674211403400.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_mask_step_pretraining_base_squadv2_epochs_3","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_mask_step_pretraining_base_squadv2_epochs_3","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_mask_step_pretraining_base_squadv2_epochs_3| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.7 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/mask_step_pretraining_roberta-base_squadv2_epochs_3 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_mask_step_pretraining_recipes_base_squadv2_epochs_3_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_mask_step_pretraining_recipes_base_squadv2_epochs_3_en.md new file mode 100644 index 00000000000000..c11df83fa70938 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_mask_step_pretraining_recipes_base_squadv2_epochs_3_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_mask_step_pretraining_recipes_base_squadv2_epochs_3 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `mask_step_pretraining_recipes-roberta-base_squadv2_epochs_3` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_mask_step_pretraining_recipes_base_squadv2_epochs_3_en_4.3.0_3.0_1674211343406.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_mask_step_pretraining_recipes_base_squadv2_epochs_3_en_4.3.0_3.0_1674211343406.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_mask_step_pretraining_recipes_base_squadv2_epochs_3","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_mask_step_pretraining_recipes_base_squadv2_epochs_3","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_mask_step_pretraining_recipes_base_squadv2_epochs_3| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|467.0 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/mask_step_pretraining_recipes-roberta-base_squadv2_epochs_3 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_mbeck_base_squad2_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_mbeck_base_squad2_en.md new file mode 100644 index 00000000000000..bc35f32a60e0da --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_mbeck_base_squad2_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from mbeck) +author: John Snow Labs +name: roberta_qa_mbeck_base_squad2 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-squad2` is a English model originally trained by `mbeck`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_mbeck_base_squad2_en_4.3.0_3.0_1674219019104.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_mbeck_base_squad2_en_4.3.0_3.0_1674219019104.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_mbeck_base_squad2","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_mbeck_base_squad2","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_mbeck_base_squad2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|459.2 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/mbeck/roberta-base-squad2 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_miamiya_base_squad2_finetuned_squad_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_miamiya_base_squad2_finetuned_squad_en.md new file mode 100644 index 00000000000000..1a5e9a34e64b08 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_miamiya_base_squad2_finetuned_squad_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from miamiya) +author: John Snow Labs +name: roberta_qa_miamiya_base_squad2_finetuned_squad +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-squad2-finetuned-squad` is a English model originally trained by `miamiya`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_miamiya_base_squad2_finetuned_squad_en_4.3.0_3.0_1674219308729.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_miamiya_base_squad2_finetuned_squad_en_4.3.0_3.0_1674219308729.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_miamiya_base_squad2_finetuned_squad","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_miamiya_base_squad2_finetuned_squad","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_miamiya_base_squad2_finetuned_squad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.3 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/miamiya/roberta-base-squad2-finetuned-squad \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_minilmv2_l6_h384_squad1.1_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_minilmv2_l6_h384_squad1.1_en.md new file mode 100644 index 00000000000000..27aac175ebf360 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_minilmv2_l6_h384_squad1.1_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Mini Cased model (from sguskin) +author: John Snow Labs +name: roberta_qa_minilmv2_l6_h384_squad1.1 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `minilmv2-L6-H384-squad1.1` is a English model originally trained by `sguskin`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_minilmv2_l6_h384_squad1.1_en_4.3.0_3.0_1674211435898.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_minilmv2_l6_h384_squad1.1_en_4.3.0_3.0_1674211435898.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_minilmv2_l6_h384_squad1.1","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_minilmv2_l6_h384_squad1.1","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_minilmv2_l6_h384_squad1.1| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|112.1 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/sguskin/minilmv2-L6-H384-squad1.1 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_model_10k_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_model_10k_en.md new file mode 100644 index 00000000000000..14685100beee43 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_model_10k_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from anablasi) +author: John Snow Labs +name: roberta_qa_model_10k +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `model_10k_qa` is a English model originally trained by `anablasi`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_model_10k_en_4.3.0_3.0_1674211482662.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_model_10k_en_4.3.0_3.0_1674211482662.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_model_10k","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_model_10k","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_model_10k| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|467.1 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/anablasi/model_10k_qa \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_model_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_model_en.md new file mode 100644 index 00000000000000..ef18638db0e341 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_model_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from choosistant) +author: John Snow Labs +name: roberta_qa_model +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `qa-model` is a English model originally trained by `choosistant`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_model_en_4.3.0_3.0_1674211832681.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_model_en_4.3.0_3.0_1674211832681.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_model","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_model","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_model| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.3 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/choosistant/qa-model \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_model_fine_tuned_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_model_fine_tuned_en.md new file mode 100644 index 00000000000000..8cafad48723d1e --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_model_fine_tuned_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from choosistant) +author: John Snow Labs +name: roberta_qa_model_fine_tuned +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `qa-model-fine-tuned` is a English model originally trained by `choosistant`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_model_fine_tuned_en_4.3.0_3.0_1674211892619.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_model_fine_tuned_en_4.3.0_3.0_1674211892619.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_model_fine_tuned","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_model_fine_tuned","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_model_fine_tuned| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.4 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/choosistant/qa-model-fine-tuned \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_modeltc_base_squad2_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_modeltc_base_squad2_en.md new file mode 100644 index 00000000000000..8153b33b54c12f --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_modeltc_base_squad2_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from ModelTC) +author: John Snow Labs +name: roberta_qa_modeltc_base_squad2 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-squad2` is a English model originally trained by `ModelTC`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_modeltc_base_squad2_en_4.3.0_3.0_1674218961578.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_modeltc_base_squad2_en_4.3.0_3.0_1674218961578.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_modeltc_base_squad2","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_modeltc_base_squad2","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_modeltc_base_squad2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.3 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/ModelTC/roberta-base-squad2 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_modeltc_base_squad_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_modeltc_base_squad_en.md new file mode 100644 index 00000000000000..e81c67ac84a641 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_modeltc_base_squad_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from ModelTC) +author: John Snow Labs +name: roberta_qa_modeltc_base_squad +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-squad` is a English model originally trained by `ModelTC`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_modeltc_base_squad_en_4.3.0_3.0_1674218615238.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_modeltc_base_squad_en_4.3.0_3.0_1674218615238.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_modeltc_base_squad","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_modeltc_base_squad","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_modeltc_base_squad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.2 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/ModelTC/roberta-base-squad \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_ms12345_base_squad2_finetuned_squad_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_ms12345_base_squad2_finetuned_squad_en.md new file mode 100644 index 00000000000000..7c8a3980a95aac --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_ms12345_base_squad2_finetuned_squad_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from ms12345) +author: John Snow Labs +name: roberta_qa_ms12345_base_squad2_finetuned_squad +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-squad2-finetuned-squad` is a English model originally trained by `ms12345`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_ms12345_base_squad2_finetuned_squad_en_4.3.0_3.0_1674219374435.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_ms12345_base_squad2_finetuned_squad_en_4.3.0_3.0_1674219374435.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_ms12345_base_squad2_finetuned_squad","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_ms12345_base_squad2_finetuned_squad","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_ms12345_base_squad2_finetuned_squad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.3 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/ms12345/roberta-base-squad2-finetuned-squad \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_msms_base_squad2_finetuned_squad_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_msms_base_squad2_finetuned_squad_en.md new file mode 100644 index 00000000000000..2b48e80d4d5713 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_msms_base_squad2_finetuned_squad_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from msms) +author: John Snow Labs +name: roberta_qa_msms_base_squad2_finetuned_squad +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-squad2-finetuned-squad` is a English model originally trained by `msms`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_msms_base_squad2_finetuned_squad_en_4.3.0_3.0_1674219434301.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_msms_base_squad2_finetuned_squad_en_4.3.0_3.0_1674219434301.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_msms_base_squad2_finetuned_squad","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_msms_base_squad2_finetuned_squad","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_msms_base_squad2_finetuned_squad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.4 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/msms/roberta-base-squad2-finetuned-squad \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_muppet_base_finetuned_squad_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_muppet_base_finetuned_squad_en.md new file mode 100644 index 00000000000000..0a45071cb2b4d9 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_muppet_base_finetuned_squad_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from janeel) +author: John Snow Labs +name: roberta_qa_muppet_base_finetuned_squad +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `muppet-roberta-base-finetuned-squad` is a English model originally trained by `janeel`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_muppet_base_finetuned_squad_en_4.3.0_3.0_1674211542810.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_muppet_base_finetuned_squad_en_4.3.0_3.0_1674211542810.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_muppet_base_finetuned_squad","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_muppet_base_finetuned_squad","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_muppet_base_finetuned_squad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.3 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/janeel/muppet-roberta-base-finetuned-squad \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_negation_detector_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_negation_detector_en.md new file mode 100644 index 00000000000000..a1cb3c91e777a2 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_negation_detector_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from Ching) +author: John Snow Labs +name: roberta_qa_negation_detector +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `negation_detector` is a English model originally trained by `Ching`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_negation_detector_en_4.3.0_3.0_1674211601485.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_negation_detector_en_4.3.0_3.0_1674211601485.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_negation_detector","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_negation_detector","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_negation_detector| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.3 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Ching/negation_detector \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_news_pretrain_ft_new_news_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_news_pretrain_ft_new_news_en.md new file mode 100644 index 00000000000000..b92edf3b8656ce --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_news_pretrain_ft_new_news_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_news_pretrain_ft_new_news +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `news_pretrain_roberta_FT_new_newsqa` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_news_pretrain_ft_new_news_en_4.3.0_3.0_1674211658822.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_news_pretrain_ft_new_news_en_4.3.0_3.0_1674211658822.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_news_pretrain_ft_new_news","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_news_pretrain_ft_new_news","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_news_pretrain_ft_new_news| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|467.1 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/news_pretrain_roberta_FT_new_newsqa \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_news_pretrain_ft_news_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_news_pretrain_ft_news_en.md new file mode 100644 index 00000000000000..24b2b7c6446066 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_news_pretrain_ft_news_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_news_pretrain_ft_news +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `news_pretrain_roberta_FT_newsqa` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_news_pretrain_ft_news_en_4.3.0_3.0_1674211715703.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_news_pretrain_ft_news_en_4.3.0_3.0_1674211715703.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_news_pretrain_ft_news","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_news_pretrain_ft_news","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_news_pretrain_ft_news| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|467.1 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/news_pretrain_roberta_FT_newsqa \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_paraphrasev3_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_paraphrasev3_en.md new file mode 100644 index 00000000000000..49768eebd6b30c --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_paraphrasev3_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from comacrae) +author: John Snow Labs +name: roberta_qa_paraphrasev3 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-paraphrasev3` is a English model originally trained by `comacrae`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_paraphrasev3_en_4.3.0_3.0_1674222313416.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_paraphrasev3_en_4.3.0_3.0_1674222313416.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_paraphrasev3","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_paraphrasev3","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_paraphrasev3| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.5 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/comacrae/roberta-paraphrasev3 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_priv_qna_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_priv_qna_en.md new file mode 100644 index 00000000000000..87cd39c54a92cb --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_priv_qna_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from arjunth2001) +author: John Snow Labs +name: roberta_qa_priv_qna +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `priv_qna` is a English model originally trained by `arjunth2001`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_priv_qna_en_4.3.0_3.0_1674211774365.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_priv_qna_en_4.3.0_3.0_1674211774365.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_priv_qna","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_priv_qna","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_priv_qna| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.6 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/arjunth2001/priv_qna \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_prk_base_squad2_finetuned_squad_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_prk_base_squad2_finetuned_squad_en.md new file mode 100644 index 00000000000000..ae1131bc7c0a1a --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_prk_base_squad2_finetuned_squad_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from prk) +author: John Snow Labs +name: roberta_qa_prk_base_squad2_finetuned_squad +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-squad2-finetuned-squad` is a English model originally trained by `prk`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_prk_base_squad2_finetuned_squad_en_4.3.0_3.0_1674219493892.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_prk_base_squad2_finetuned_squad_en_4.3.0_3.0_1674219493892.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_prk_base_squad2_finetuned_squad","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_prk_base_squad2_finetuned_squad","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_prk_base_squad2_finetuned_squad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.3 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/prk/roberta-base-squad2-finetuned-squad \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_quales_iberlef_squad_2_es.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_quales_iberlef_squad_2_es.md new file mode 100644 index 00000000000000..3d6916bf0a0a41 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_quales_iberlef_squad_2_es.md @@ -0,0 +1,88 @@ +--- +layout: model +title: Spanish RobertaForQuestionAnswering Cased model (from stevemobs) +author: John Snow Labs +name: roberta_qa_quales_iberlef_squad_2 +date: 2023-01-20 +tags: [es, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: es +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `quales-iberlef-squad_2` is a Spanish model originally trained by `stevemobs`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_quales_iberlef_squad_2_es_4.3.0_3.0_1674212017084.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_quales_iberlef_squad_2_es_4.3.0_3.0_1674212017084.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_quales_iberlef_squad_2","es")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_quales_iberlef_squad_2","es") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_quales_iberlef_squad_2| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|es| +|Size:|1.3 GB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/stevemobs/quales-iberlef-squad_2 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rahulchakwate_base_finetuned_squad_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rahulchakwate_base_finetuned_squad_en.md new file mode 100644 index 00000000000000..f239f40e152cc7 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rahulchakwate_base_finetuned_squad_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from rahulchakwate) +author: John Snow Labs +name: roberta_qa_rahulchakwate_base_finetuned_squad +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-base-finetuned-squad` is a English model originally trained by `rahulchakwate`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_rahulchakwate_base_finetuned_squad_en_4.3.0_3.0_1674217478490.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_rahulchakwate_base_finetuned_squad_en_4.3.0_3.0_1674217478490.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_rahulchakwate_base_finetuned_squad","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_rahulchakwate_base_finetuned_squad","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_rahulchakwate_base_finetuned_squad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.3 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/rahulchakwate/roberta-base-finetuned-squad \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_re_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_re_en.md new file mode 100644 index 00000000000000..f54b52bb773ac1 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_re_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from shmuelamar) +author: John Snow Labs +name: roberta_qa_re +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `REQA-RoBERTa` is a English model originally trained by `shmuelamar`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_re_en_4.3.0_3.0_1674208450623.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_re_en_4.3.0_3.0_1674208450623.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_re","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_re","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_re| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|1.3 GB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/shmuelamar/REQA-RoBERTa \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_recipe_triplet_recipes_base_easy_squadv2_epochs_3_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_recipe_triplet_recipes_base_easy_squadv2_epochs_3_en.md new file mode 100644 index 00000000000000..2f649df95025c6 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_recipe_triplet_recipes_base_easy_squadv2_epochs_3_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_recipe_triplet_recipes_base_easy_squadv2_epochs_3 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `recipe_triplet_recipes-roberta-base_EASY_squadv2_epochs_3` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_recipe_triplet_recipes_base_easy_squadv2_epochs_3_en_4.3.0_3.0_1674212163630.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_recipe_triplet_recipes_base_easy_squadv2_epochs_3_en_4.3.0_3.0_1674212163630.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_recipe_triplet_recipes_base_easy_squadv2_epochs_3","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_recipe_triplet_recipes_base_easy_squadv2_epochs_3","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_recipe_triplet_recipes_base_easy_squadv2_epochs_3| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|467.0 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/recipe_triplet_recipes-roberta-base_EASY_squadv2_epochs_3 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_recipe_triplet_recipes_base_easy_timestep_squadv2_epochs_3_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_recipe_triplet_recipes_base_easy_timestep_squadv2_epochs_3_en.md new file mode 100644 index 00000000000000..2309758f6f0e02 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_recipe_triplet_recipes_base_easy_timestep_squadv2_epochs_3_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_recipe_triplet_recipes_base_easy_timestep_squadv2_epochs_3 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `recipe_triplet_recipes-roberta-base_EASY_TIMESTEP_squadv2_epochs_3` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_recipe_triplet_recipes_base_easy_timestep_squadv2_epochs_3_en_4.3.0_3.0_1674212108351.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_recipe_triplet_recipes_base_easy_timestep_squadv2_epochs_3_en_4.3.0_3.0_1674212108351.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_recipe_triplet_recipes_base_easy_timestep_squadv2_epochs_3","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_recipe_triplet_recipes_base_easy_timestep_squadv2_epochs_3","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_recipe_triplet_recipes_base_easy_timestep_squadv2_epochs_3| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|467.0 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/recipe_triplet_recipes-roberta-base_EASY_TIMESTEP_squadv2_epochs_3 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_recipe_triplet_recipes_base_squadv2_epochs_3_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_recipe_triplet_recipes_base_squadv2_epochs_3_en.md new file mode 100644 index 00000000000000..5bcbf549a784f6 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_recipe_triplet_recipes_base_squadv2_epochs_3_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_recipe_triplet_recipes_base_squadv2_epochs_3 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `recipe_triplet_recipes-roberta-base_squadv2_epochs_3` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_recipe_triplet_recipes_base_squadv2_epochs_3_en_4.3.0_3.0_1674212279557.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_recipe_triplet_recipes_base_squadv2_epochs_3_en_4.3.0_3.0_1674212279557.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_recipe_triplet_recipes_base_squadv2_epochs_3","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_recipe_triplet_recipes_base_squadv2_epochs_3","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_recipe_triplet_recipes_base_squadv2_epochs_3| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|467.0 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/recipe_triplet_recipes-roberta-base_squadv2_epochs_3 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_recipe_triplet_recipes_base_timestep_squadv2_epochs_3_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_recipe_triplet_recipes_base_timestep_squadv2_epochs_3_en.md new file mode 100644 index 00000000000000..f33701da4c0b71 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_recipe_triplet_recipes_base_timestep_squadv2_epochs_3_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_recipe_triplet_recipes_base_timestep_squadv2_epochs_3 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `recipe_triplet_recipes-roberta-base_TIMESTEP_squadv2_epochs_3` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_recipe_triplet_recipes_base_timestep_squadv2_epochs_3_en_4.3.0_3.0_1674212220798.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_recipe_triplet_recipes_base_timestep_squadv2_epochs_3_en_4.3.0_3.0_1674212220798.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_recipe_triplet_recipes_base_timestep_squadv2_epochs_3","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_recipe_triplet_recipes_base_timestep_squadv2_epochs_3","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_recipe_triplet_recipes_base_timestep_squadv2_epochs_3| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|467.0 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/recipe_triplet_recipes-roberta-base_TIMESTEP_squadv2_epochs_3 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_robbert_base_squad_finetuned_on_runaways_nl.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_robbert_base_squad_finetuned_on_runaways_nl.md new file mode 100644 index 00000000000000..838a8c587f46ec --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_robbert_base_squad_finetuned_on_runaways_nl.md @@ -0,0 +1,88 @@ +--- +layout: model +title: Dutch RobertaForQuestionAnswering Base Cased model (from Nadav) +author: John Snow Labs +name: roberta_qa_robbert_base_squad_finetuned_on_runaways +date: 2023-01-20 +tags: [nl, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: nl +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `robbert-base-squad-finetuned-on-runaways-nl` is a Dutch model originally trained by `Nadav`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_robbert_base_squad_finetuned_on_runaways_nl_4.3.0_3.0_1674212455477.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_robbert_base_squad_finetuned_on_runaways_nl_4.3.0_3.0_1674212455477.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_robbert_base_squad_finetuned_on_runaways","nl")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_robbert_base_squad_finetuned_on_runaways","nl") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_robbert_base_squad_finetuned_on_runaways| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|nl| +|Size:|436.4 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Nadav/robbert-base-squad-finetuned-on-runaways-nl \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_roberta_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_roberta_en.md new file mode 100644 index 00000000000000..9ea7a2be785e2e --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_roberta_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from nlpunibo) +author: John Snow Labs +name: roberta_qa_roberta +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta` is a English model originally trained by `nlpunibo`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_roberta_en_4.3.0_3.0_1674212513043.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_roberta_en_4.3.0_3.0_1674212513043.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_roberta","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_roberta","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_roberta| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|463.6 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/nlpunibo/roberta \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_robertaabsa_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_robertaabsa_en.md new file mode 100644 index 00000000000000..e49770aa709316 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_robertaabsa_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from LucasS) +author: John Snow Labs +name: roberta_qa_robertaabsa +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `robertaABSA` is a English model originally trained by `LucasS`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_robertaabsa_en_4.3.0_3.0_1674222776379.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_robertaabsa_en_4.3.0_3.0_1674222776379.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_robertaabsa","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_robertaabsa","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_robertaabsa| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|437.1 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/LucasS/robertaABSA \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_robertabaseabsa_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_robertabaseabsa_en.md new file mode 100644 index 00000000000000..caa0b1e1907faa --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_robertabaseabsa_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from LucasS) +author: John Snow Labs +name: roberta_qa_robertabaseabsa +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `robertaBaseABSA` is a English model originally trained by `LucasS`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_robertabaseabsa_en_4.3.0_3.0_1674222849343.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_robertabaseabsa_en_4.3.0_3.0_1674222849343.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_robertabaseabsa","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_robertabaseabsa","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_robertabaseabsa| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|437.1 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/LucasS/robertaBaseABSA \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_bert_quadruplet_epochs_1_shard_1_squad2.0_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_bert_quadruplet_epochs_1_shard_1_squad2.0_en.md new file mode 100644 index 00000000000000..2189a6d9b99d87 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_bert_quadruplet_epochs_1_shard_1_squad2.0_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_rule_based_bert_quadruplet_epochs_1_shard_1_squad2.0 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `rule_based_roberta_bert_quadruplet_epochs_1_shard_1_squad2.0` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_rule_based_bert_quadruplet_epochs_1_shard_1_squad2.0_en_4.3.0_3.0_1674223261564.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_rule_based_bert_quadruplet_epochs_1_shard_1_squad2.0_en_4.3.0_3.0_1674223261564.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_rule_based_bert_quadruplet_epochs_1_shard_1_squad2.0","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_rule_based_bert_quadruplet_epochs_1_shard_1_squad2.0","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_rule_based_bert_quadruplet_epochs_1_shard_1_squad2.0| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|460.4 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/rule_based_roberta_bert_quadruplet_epochs_1_shard_1_squad2.0 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_bert_triplet_epochs_1_shard_1_squad2.0_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_bert_triplet_epochs_1_shard_1_squad2.0_en.md new file mode 100644 index 00000000000000..1c01a3a5923114 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_bert_triplet_epochs_1_shard_1_squad2.0_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_rule_based_bert_triplet_epochs_1_shard_1_squad2.0 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `rule_based_roberta_bert_triplet_epochs_1_shard_1_squad2.0` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_rule_based_bert_triplet_epochs_1_shard_1_squad2.0_en_4.3.0_3.0_1674223322432.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_rule_based_bert_triplet_epochs_1_shard_1_squad2.0_en_4.3.0_3.0_1674223322432.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_rule_based_bert_triplet_epochs_1_shard_1_squad2.0","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_rule_based_bert_triplet_epochs_1_shard_1_squad2.0","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_rule_based_bert_triplet_epochs_1_shard_1_squad2.0| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|460.5 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/rule_based_roberta_bert_triplet_epochs_1_shard_1_squad2.0 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_hier_quadruplet_0.1_epochs_1_shard_1_squad2.0_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_hier_quadruplet_0.1_epochs_1_shard_1_squad2.0_en.md new file mode 100644 index 00000000000000..079ed5c752b8a9 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_hier_quadruplet_0.1_epochs_1_shard_1_squad2.0_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_rule_based_hier_quadruplet_0.1_epochs_1_shard_1_squad2.0 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `rule_based_roberta_hier_quadruplet_0.1_epochs_1_shard_1_squad2.0` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_rule_based_hier_quadruplet_0.1_epochs_1_shard_1_squad2.0_en_4.3.0_3.0_1674223386005.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_rule_based_hier_quadruplet_0.1_epochs_1_shard_1_squad2.0_en_4.3.0_3.0_1674223386005.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_rule_based_hier_quadruplet_0.1_epochs_1_shard_1_squad2.0","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_rule_based_hier_quadruplet_0.1_epochs_1_shard_1_squad2.0","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_rule_based_hier_quadruplet_0.1_epochs_1_shard_1_squad2.0| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|460.4 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/rule_based_roberta_hier_quadruplet_0.1_epochs_1_shard_1_squad2.0 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_hier_quadruplet_epochs_1_shard_1_squad2.0_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_hier_quadruplet_epochs_1_shard_1_squad2.0_en.md new file mode 100644 index 00000000000000..eae74358dca7b6 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_hier_quadruplet_epochs_1_shard_1_squad2.0_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_rule_based_hier_quadruplet_epochs_1_shard_1_squad2.0 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `rule_based_roberta_hier_quadruplet_epochs_1_shard_1_squad2.0` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_rule_based_hier_quadruplet_epochs_1_shard_1_squad2.0_en_4.3.0_3.0_1674223448939.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_rule_based_hier_quadruplet_epochs_1_shard_1_squad2.0_en_4.3.0_3.0_1674223448939.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_rule_based_hier_quadruplet_epochs_1_shard_1_squad2.0","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_rule_based_hier_quadruplet_epochs_1_shard_1_squad2.0","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_rule_based_hier_quadruplet_epochs_1_shard_1_squad2.0| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|460.4 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/rule_based_roberta_hier_quadruplet_epochs_1_shard_1_squad2.0 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_hier_triplet_0.1_epochs_1_shard_1_squad2.0_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_hier_triplet_0.1_epochs_1_shard_1_squad2.0_en.md new file mode 100644 index 00000000000000..b0192027ba1c8f --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_hier_triplet_0.1_epochs_1_shard_1_squad2.0_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_rule_based_hier_triplet_0.1_epochs_1_shard_1_squad2.0 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `rule_based_roberta_hier_triplet_0.1_epochs_1_shard_1_squad2.0` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_rule_based_hier_triplet_0.1_epochs_1_shard_1_squad2.0_en_4.3.0_3.0_1674223509314.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_rule_based_hier_triplet_0.1_epochs_1_shard_1_squad2.0_en_4.3.0_3.0_1674223509314.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_rule_based_hier_triplet_0.1_epochs_1_shard_1_squad2.0","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_rule_based_hier_triplet_0.1_epochs_1_shard_1_squad2.0","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_rule_based_hier_triplet_0.1_epochs_1_shard_1_squad2.0| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|460.4 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/rule_based_roberta_hier_triplet_0.1_epochs_1_shard_1_squad2.0 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_hier_triplet_epochs_1_shard_1_squad2.0_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_hier_triplet_epochs_1_shard_1_squad2.0_en.md new file mode 100644 index 00000000000000..0bce0e66746ad5 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_hier_triplet_epochs_1_shard_1_squad2.0_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_rule_based_hier_triplet_epochs_1_shard_1_squad2.0 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `rule_based_roberta_hier_triplet_epochs_1_shard_1_squad2.0` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_rule_based_hier_triplet_epochs_1_shard_1_squad2.0_en_4.3.0_3.0_1674223579788.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_rule_based_hier_triplet_epochs_1_shard_1_squad2.0_en_4.3.0_3.0_1674223579788.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_rule_based_hier_triplet_epochs_1_shard_1_squad2.0","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_rule_based_hier_triplet_epochs_1_shard_1_squad2.0","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_rule_based_hier_triplet_epochs_1_shard_1_squad2.0| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|460.4 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/rule_based_roberta_hier_triplet_epochs_1_shard_1_squad2.0 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_hier_triplet_shuffled_epochs_1_shard_1_squad2.0_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_hier_triplet_shuffled_epochs_1_shard_1_squad2.0_en.md new file mode 100644 index 00000000000000..a993ab45236986 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_hier_triplet_shuffled_epochs_1_shard_1_squad2.0_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_rule_based_hier_triplet_shuffled_epochs_1_shard_1_squad2.0 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `rule_based_roberta_hier_triplet_shuffled_epochs_1_shard_1_squad2.0` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_rule_based_hier_triplet_shuffled_epochs_1_shard_1_squad2.0_en_4.3.0_3.0_1674223641296.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_rule_based_hier_triplet_shuffled_epochs_1_shard_1_squad2.0_en_4.3.0_3.0_1674223641296.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_rule_based_hier_triplet_shuffled_epochs_1_shard_1_squad2.0","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_rule_based_hier_triplet_shuffled_epochs_1_shard_1_squad2.0","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_rule_based_hier_triplet_shuffled_epochs_1_shard_1_squad2.0| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|460.5 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/rule_based_roberta_hier_triplet_shuffled_epochs_1_shard_1_squad2.0 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_hier_triplet_shuffled_paras_epochs_1_shard_1_squad2.0_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_hier_triplet_shuffled_paras_epochs_1_shard_1_squad2.0_en.md new file mode 100644 index 00000000000000..9917bb9dcc8725 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_hier_triplet_shuffled_paras_epochs_1_shard_1_squad2.0_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_rule_based_hier_triplet_shuffled_paras_epochs_1_shard_1_squad2.0 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `rule_based_roberta_hier_triplet_shuffled_paras_epochs_1_shard_1_squad2.0` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_rule_based_hier_triplet_shuffled_paras_epochs_1_shard_1_squad2.0_en_4.3.0_3.0_1674223704448.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_rule_based_hier_triplet_shuffled_paras_epochs_1_shard_1_squad2.0_en_4.3.0_3.0_1674223704448.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_rule_based_hier_triplet_shuffled_paras_epochs_1_shard_1_squad2.0","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_rule_based_hier_triplet_shuffled_paras_epochs_1_shard_1_squad2.0","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_rule_based_hier_triplet_shuffled_paras_epochs_1_shard_1_squad2.0| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|460.4 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/rule_based_roberta_hier_triplet_shuffled_paras_epochs_1_shard_1_squad2.0 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_only_classfn_epochs_1_shard_1_squad2.0_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_only_classfn_epochs_1_shard_1_squad2.0_en.md new file mode 100644 index 00000000000000..4fed39700c5f9c --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_only_classfn_epochs_1_shard_1_squad2.0_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_rule_based_only_classfn_epochs_1_shard_1_squad2.0 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `rule_based_roberta_only_classfn_epochs_1_shard_1_squad2.0` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_rule_based_only_classfn_epochs_1_shard_1_squad2.0_en_4.3.0_3.0_1674223765535.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_rule_based_only_classfn_epochs_1_shard_1_squad2.0_en_4.3.0_3.0_1674223765535.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_rule_based_only_classfn_epochs_1_shard_1_squad2.0","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_rule_based_only_classfn_epochs_1_shard_1_squad2.0","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_rule_based_only_classfn_epochs_1_shard_1_squad2.0| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|460.4 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/rule_based_roberta_only_classfn_epochs_1_shard_1_squad2.0 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_only_classfn_twostage_epochs_1_shard_1_squad2.0_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_only_classfn_twostage_epochs_1_shard_1_squad2.0_en.md new file mode 100644 index 00000000000000..b822be0e06dedb --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_only_classfn_twostage_epochs_1_shard_1_squad2.0_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_rule_based_only_classfn_twostage_epochs_1_shard_1_squad2.0 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `rule_based_roberta_only_classfn_twostage_epochs_1_shard_1_squad2.0` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_rule_based_only_classfn_twostage_epochs_1_shard_1_squad2.0_en_4.3.0_3.0_1674223837089.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_rule_based_only_classfn_twostage_epochs_1_shard_1_squad2.0_en_4.3.0_3.0_1674223837089.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_rule_based_only_classfn_twostage_epochs_1_shard_1_squad2.0","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_rule_based_only_classfn_twostage_epochs_1_shard_1_squad2.0","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_rule_based_only_classfn_twostage_epochs_1_shard_1_squad2.0| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|463.4 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/rule_based_roberta_only_classfn_twostage_epochs_1_shard_1_squad2.0 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_twostage_quadruplet_epochs_1_shard_1_squad2.0_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_twostage_quadruplet_epochs_1_shard_1_squad2.0_en.md new file mode 100644 index 00000000000000..64b2bc517264d8 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_twostage_quadruplet_epochs_1_shard_1_squad2.0_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_rule_based_twostage_quadruplet_epochs_1_shard_1_squad2.0 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `rule_based_roberta_twostage_quadruplet_epochs_1_shard_1_squad2.0` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_rule_based_twostage_quadruplet_epochs_1_shard_1_squad2.0_en_4.3.0_3.0_1674223883289.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_rule_based_twostage_quadruplet_epochs_1_shard_1_squad2.0_en_4.3.0_3.0_1674223883289.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_rule_based_twostage_quadruplet_epochs_1_shard_1_squad2.0","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_rule_based_twostage_quadruplet_epochs_1_shard_1_squad2.0","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_rule_based_twostage_quadruplet_epochs_1_shard_1_squad2.0| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|306.8 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/rule_based_roberta_twostage_quadruplet_epochs_1_shard_1_squad2.0 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_twostagequadruplet_hier_epochs_1_shard_1_squad2.0_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_twostagequadruplet_hier_epochs_1_shard_1_squad2.0_en.md new file mode 100644 index 00000000000000..e3f560d6e4ecda --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_twostagequadruplet_hier_epochs_1_shard_1_squad2.0_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_rule_based_twostagequadruplet_hier_epochs_1_shard_1_squad2.0 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `rule_based_roberta_twostagequadruplet_hier_epochs_1_shard_1_squad2.0` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_rule_based_twostagequadruplet_hier_epochs_1_shard_1_squad2.0_en_4.3.0_3.0_1674223924983.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_rule_based_twostagequadruplet_hier_epochs_1_shard_1_squad2.0_en_4.3.0_3.0_1674223924983.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_rule_based_twostagequadruplet_hier_epochs_1_shard_1_squad2.0","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_rule_based_twostagequadruplet_hier_epochs_1_shard_1_squad2.0","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_rule_based_twostagequadruplet_hier_epochs_1_shard_1_squad2.0| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|306.8 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/rule_based_roberta_twostagequadruplet_hier_epochs_1_shard_1_squad2.0 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_twostagetriplet_epochs_1_shard_1_squad2.0_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_twostagetriplet_epochs_1_shard_1_squad2.0_en.md new file mode 100644 index 00000000000000..a0dc484ae5211f --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_twostagetriplet_epochs_1_shard_1_squad2.0_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_rule_based_twostagetriplet_epochs_1_shard_1_squad2.0 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `rule_based_roberta_twostagetriplet_epochs_1_shard_1_squad2.0` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_rule_based_twostagetriplet_epochs_1_shard_1_squad2.0_en_4.3.0_3.0_1674223967265.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_rule_based_twostagetriplet_epochs_1_shard_1_squad2.0_en_4.3.0_3.0_1674223967265.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_rule_based_twostagetriplet_epochs_1_shard_1_squad2.0","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_rule_based_twostagetriplet_epochs_1_shard_1_squad2.0","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_rule_based_twostagetriplet_epochs_1_shard_1_squad2.0| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|306.8 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/rule_based_roberta_twostagetriplet_epochs_1_shard_1_squad2.0 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_twostagetriplet_hier_epochs_1_shard_1_squad2.0_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_twostagetriplet_hier_epochs_1_shard_1_squad2.0_en.md new file mode 100644 index 00000000000000..48ff9cb3be8748 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_rule_based_twostagetriplet_hier_epochs_1_shard_1_squad2.0_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_rule_based_twostagetriplet_hier_epochs_1_shard_1_squad2.0 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `rule_based_roberta_twostagetriplet_hier_epochs_1_shard_1_squad2.0` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_rule_based_twostagetriplet_hier_epochs_1_shard_1_squad2.0_en_4.3.0_3.0_1674224008013.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_rule_based_twostagetriplet_hier_epochs_1_shard_1_squad2.0_en_4.3.0_3.0_1674224008013.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_rule_based_twostagetriplet_hier_epochs_1_shard_1_squad2.0","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_rule_based_twostagetriplet_hier_epochs_1_shard_1_squad2.0","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_rule_based_twostagetriplet_hier_epochs_1_shard_1_squad2.0| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|306.8 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/rule_based_roberta_twostagetriplet_hier_epochs_1_shard_1_squad2.0 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_sae_base_squad_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_sae_base_squad_en.md new file mode 100644 index 00000000000000..6bd2aacc0a3aa9 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_sae_base_squad_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from jgammack) +author: John Snow Labs +name: roberta_qa_sae_base_squad +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `SAE-roberta-base-squad` is a English model originally trained by `jgammack`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_sae_base_squad_en_4.3.0_3.0_1674208844842.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_sae_base_squad_en_4.3.0_3.0_1674208844842.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_sae_base_squad","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_sae_base_squad","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_sae_base_squad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|467.1 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/jgammack/SAE-roberta-base-squad \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_squad_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_squad_en.md new file mode 100644 index 00000000000000..be7582a9270407 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_squad_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from Raynok) +author: John Snow Labs +name: roberta_qa_squad +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-squad` is a English model originally trained by `Raynok`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_squad_en_4.3.0_3.0_1674222368310.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_squad_en_4.3.0_3.0_1674222368310.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_squad","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_squad","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_squad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.2 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Raynok/roberta-squad \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_squadv2_base_3_epochs_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_squadv2_base_3_epochs_en.md new file mode 100644 index 00000000000000..8cdd8a8df84f82 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_squadv2_base_3_epochs_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_squadv2_base_3_epochs +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `squadv2-roberta-base-3-epochs` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_squadv2_base_3_epochs_en_4.3.0_3.0_1674224185355.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_squadv2_base_3_epochs_en_4.3.0_3.0_1674224185355.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_squadv2_base_3_epochs","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_squadv2_base_3_epochs","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_squadv2_base_3_epochs| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|460.5 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/squadv2-roberta-base-3-epochs \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_squadv2_recipe_3_epochs_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_squadv2_recipe_3_epochs_en.md new file mode 100644 index 00000000000000..0016e8f8b3e409 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_squadv2_recipe_3_epochs_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_squadv2_recipe_3_epochs +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `squadv2-recipe-roberta-3-epochs` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_squadv2_recipe_3_epochs_en_4.3.0_3.0_1674224063985.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_squadv2_recipe_3_epochs_en_4.3.0_3.0_1674224063985.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_squadv2_recipe_3_epochs","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_squadv2_recipe_3_epochs","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_squadv2_recipe_3_epochs| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|467.0 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/squadv2-recipe-roberta-3-epochs \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_squadv2_recipe_tokenwise_token_and_step_losses_3_epochs_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_squadv2_recipe_tokenwise_token_and_step_losses_3_epochs_en.md new file mode 100644 index 00000000000000..8255b8e70b3f0e --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_squadv2_recipe_tokenwise_token_and_step_losses_3_epochs_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from AnonymousSub) +author: John Snow Labs +name: roberta_qa_squadv2_recipe_tokenwise_token_and_step_losses_3_epochs +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `squadv2-recipe-roberta-tokenwise-token-and-step-losses-3-epochs` is a English model originally trained by `AnonymousSub`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_squadv2_recipe_tokenwise_token_and_step_losses_3_epochs_en_4.3.0_3.0_1674224122519.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_squadv2_recipe_tokenwise_token_and_step_losses_3_epochs_en_4.3.0_3.0_1674224122519.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_squadv2_recipe_tokenwise_token_and_step_losses_3_epochs","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_squadv2_recipe_tokenwise_token_and_step_losses_3_epochs","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_squadv2_recipe_tokenwise_token_and_step_losses_3_epochs| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|467.1 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/AnonymousSub/squadv2-recipe-roberta-tokenwise-token-and-step-losses-3-epochs \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_test_v1_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_test_v1_en.md new file mode 100644 index 00000000000000..adeb0348ab963d --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_test_v1_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from Andranik) +author: John Snow Labs +name: roberta_qa_test_v1 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `TestQaV1` is a English model originally trained by `Andranik`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_test_v1_en_4.3.0_3.0_1674208905005.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_test_v1_en_4.3.0_3.0_1674208905005.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_test_v1","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_test_v1","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_test_v1| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.3 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/Andranik/TestQaV1 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_testabsa3_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_testabsa3_en.md new file mode 100644 index 00000000000000..bbcde0ad42e707 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_testabsa3_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from eAsyle) +author: John Snow Labs +name: roberta_qa_testabsa3 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `testABSA3` is a English model originally trained by `eAsyle`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_testabsa3_en_4.3.0_3.0_1674224346874.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_testabsa3_en_4.3.0_3.0_1674224346874.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_testabsa3","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_testabsa3","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_testabsa3| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|426.2 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/eAsyle/testABSA3 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_testabsa_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_testabsa_en.md new file mode 100644 index 00000000000000..a90489e108cf75 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_testabsa_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from eAsyle) +author: John Snow Labs +name: roberta_qa_testabsa +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `testABSA` is a English model originally trained by `eAsyle`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_testabsa_en_4.3.0_3.0_1674224267018.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_testabsa_en_4.3.0_3.0_1674224267018.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_testabsa","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_testabsa","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_testabsa| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|426.2 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/eAsyle/testABSA \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_tiny_random_forquestionanswering_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_tiny_random_forquestionanswering_en.md new file mode 100644 index 00000000000000..a9cabd1718f630 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_tiny_random_forquestionanswering_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Tiny Cased model (from hf-internal-testing) +author: John Snow Labs +name: roberta_qa_tiny_random_forquestionanswering +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `tiny-random-RobertaForQuestionAnswering` is a English model originally trained by `hf-internal-testing`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_tiny_random_forquestionanswering_en_4.3.0_3.0_1674224369695.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_tiny_random_forquestionanswering_en_4.3.0_3.0_1674224369695.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_tiny_random_forquestionanswering","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_tiny_random_forquestionanswering","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_tiny_random_forquestionanswering| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|681.7 KB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/hf-internal-testing/tiny-random-RobertaForQuestionAnswering \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_tiny_squad2_step1_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_tiny_squad2_step1_en.md new file mode 100644 index 00000000000000..836a32d34b2cf2 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_tiny_squad2_step1_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Tiny Cased model (from deepset) +author: John Snow Labs +name: roberta_qa_tiny_squad2_step1 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `tinyroberta-squad2-step1` is a English model originally trained by `deepset`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_tiny_squad2_step1_en_4.3.0_3.0_1674224441422.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_tiny_squad2_step1_en_4.3.0_3.0_1674224441422.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_tiny_squad2_step1","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_tiny_squad2_step1","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_tiny_squad2_step1| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|307.3 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/deepset/tinyroberta-squad2-step1 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_train_json_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_train_json_en.md new file mode 100644 index 00000000000000..dcdccc1bf21f46 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_train_json_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from aravind-812) +author: John Snow Labs +name: roberta_qa_train_json +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-train-json` is a English model originally trained by `aravind-812`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_train_json_en_4.3.0_3.0_1674222495676.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_train_json_en_4.3.0_3.0_1674222495676.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_train_json","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_train_json","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_train_json| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|1.3 GB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/aravind-812/roberta-train-json \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_tydi_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_tydi_en.md new file mode 100644 index 00000000000000..29f900c0653fbf --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_tydi_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from benny6) +author: John Snow Labs +name: roberta_qa_tydi +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-tydiqa` is a English model originally trained by `benny6`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_tydi_en_4.3.0_3.0_1674222584111.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_tydi_en_4.3.0_3.0_1674222584111.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_tydi","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_tydi","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_tydi| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|471.7 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/benny6/roberta-tydiqa \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_unaugmentedv3_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_unaugmentedv3_en.md new file mode 100644 index 00000000000000..b257468d6363e3 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_unaugmentedv3_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from comacrae) +author: John Snow Labs +name: roberta_qa_unaugmentedv3 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-unaugmentedv3` is a English model originally trained by `comacrae`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_unaugmentedv3_en_4.3.0_3.0_1674222641663.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_unaugmentedv3_en_4.3.0_3.0_1674222641663.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_unaugmentedv3","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_unaugmentedv3","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_unaugmentedv3| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.5 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/comacrae/roberta-unaugmentedv3 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_unaugv3_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_unaugv3_en.md new file mode 100644 index 00000000000000..4e3ef6b86eaa69 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_unaugv3_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Cased model (from comacrae) +author: John Snow Labs +name: roberta_qa_unaugv3 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `roberta-unaugv3` is a English model originally trained by `comacrae`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_unaugv3_en_4.3.0_3.0_1674222699716.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_unaugv3_en_4.3.0_3.0_1674222699716.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_unaugv3","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_unaugv3","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_unaugv3| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|464.5 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/comacrae/roberta-unaugv3 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_unqover_base_news_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_unqover_base_news_en.md new file mode 100644 index 00000000000000..3d7b8b119a39cb --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_unqover_base_news_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from tli8hf) +author: John Snow Labs +name: roberta_qa_unqover_base_news +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `unqover-roberta-base-newsqa` is a English model originally trained by `tli8hf`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_unqover_base_news_en_4.3.0_3.0_1674224491216.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_unqover_base_news_en_4.3.0_3.0_1674224491216.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_unqover_base_news","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_unqover_base_news","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_unqover_base_news| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|463.7 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/tli8hf/unqover-roberta-base-newsqa \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_unqover_base_squad_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_unqover_base_squad_en.md new file mode 100644 index 00000000000000..ae8d5bd136243a --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_unqover_base_squad_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from tli8hf) +author: John Snow Labs +name: roberta_qa_unqover_base_squad +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `unqover-roberta-base-squad` is a English model originally trained by `tli8hf`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_unqover_base_squad_en_4.3.0_3.0_1674224552223.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_unqover_base_squad_en_4.3.0_3.0_1674224552223.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_unqover_base_squad","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_unqover_base_squad","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_unqover_base_squad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|463.8 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/tli8hf/unqover-roberta-base-squad \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_unqover_large_news_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_unqover_large_news_en.md new file mode 100644 index 00000000000000..1a3db4011e89de --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_unqover_large_news_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Large Cased model (from tli8hf) +author: John Snow Labs +name: roberta_qa_unqover_large_news +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `unqover-roberta-large-newsqa` is a English model originally trained by `tli8hf`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_unqover_large_news_en_4.3.0_3.0_1674224676431.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_unqover_large_news_en_4.3.0_3.0_1674224676431.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_unqover_large_news","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_unqover_large_news","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_unqover_large_news| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|1.3 GB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/tli8hf/unqover-roberta-large-newsqa \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_unqover_large_squad_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_unqover_large_squad_en.md new file mode 100644 index 00000000000000..4bcced4f018064 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_unqover_large_squad_en.md @@ -0,0 +1,88 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Large Cased model (from tli8hf) +author: John Snow Labs +name: roberta_qa_unqover_large_squad +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `unqover-roberta-large-squad` is a English model originally trained by `tli8hf`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_unqover_large_squad_en_4.3.0_3.0_1674224834331.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_unqover_large_squad_en_4.3.0_3.0_1674224834331.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_unqover_large_squad","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_unqover_large_squad","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_unqover_large_squad| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|1.3 GB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/tli8hf/unqover-roberta-large-squad \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_xdoc_base_squad1.1_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_xdoc_base_squad1.1_en.md new file mode 100644 index 00000000000000..9f31702ceb6528 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_xdoc_base_squad1.1_en.md @@ -0,0 +1,89 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from microsoft) +author: John Snow Labs +name: roberta_qa_xdoc_base_squad1.1 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `xdoc-base-squad1.1` is a English model originally trained by `microsoft`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_xdoc_base_squad1.1_en_4.3.0_3.0_1674224925472.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_xdoc_base_squad1.1_en_4.3.0_3.0_1674224925472.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_xdoc_base_squad1.1","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_xdoc_base_squad1.1","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_xdoc_base_squad1.1| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|466.9 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/microsoft/xdoc-base-squad1.1 +- https://arxiv.org/abs/2210.02849 \ No newline at end of file diff --git a/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_xdoc_base_squad2.0_en.md b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_xdoc_base_squad2.0_en.md new file mode 100644 index 00000000000000..a9dfd237dfc675 --- /dev/null +++ b/docs/_posts/Damla-Gurbaz/2023-01-20-roberta_qa_xdoc_base_squad2.0_en.md @@ -0,0 +1,89 @@ +--- +layout: model +title: English RobertaForQuestionAnswering Base Cased model (from microsoft) +author: John Snow Labs +name: roberta_qa_xdoc_base_squad2.0 +date: 2023-01-20 +tags: [en, open_source, roberta, question_answering, tensorflow] +task: Question Answering +language: en +edition: Spark NLP 4.3.0 +spark_version: 3.0 +supported: true +engine: tensorflow +annotator: RoBertaForQuestionAnswering +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained RobertaForQuestionAnswering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `xdoc-base-squad2.0` is a English model originally trained by `microsoft`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/roberta_qa_xdoc_base_squad2.0_en_4.3.0_3.0_1674224984469.zip){:.button.button-orange} +[Copy S3 URI](s3://auxdata.johnsnowlabs.com/public/models/roberta_qa_xdoc_base_squad2.0_en_4.3.0_3.0_1674224984469.zip){:.button.button-orange.button-orange-trans.button-icon.button-copy-s3} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +Document_Assembler = MultiDocumentAssembler()\ + .setInputCols(["question", "context"])\ + .setOutputCols(["document_question", "document_context"]) + +Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_xdoc_base_squad2.0","en")\ + .setInputCols(["document_question", "document_context"])\ + .setOutputCol("answer")\ + .setCaseSensitive(True) + +pipeline = Pipeline(stages=[Document_Assembler, Question_Answering]) + +data = spark.createDataFrame([["What's my name?","My name is Clara and I live in Berkeley."]]).toDF("question", "context") + +result = pipeline.fit(data).transform(data) +``` +```scala +val Document_Assembler = new MultiDocumentAssembler() + .setInputCols(Array("question", "context")) + .setOutputCols(Array("document_question", "document_context")) + +val Question_Answering = RoBertaForQuestionAnswering.pretrained("roberta_qa_xdoc_base_squad2.0","en") + .setInputCols(Array("document_question", "document_context")) + .setOutputCol("answer") + .setCaseSensitive(true) + +val pipeline = new Pipeline().setStages(Array(Document_Assembler, Question_Answering)) + +val data = Seq("What's my name?","My name is Clara and I live in Berkeley.").toDS.toDF("question", "context") + +val result = pipeline.fit(data).transform(data) +``` +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|roberta_qa_xdoc_base_squad2.0| +|Compatibility:|Spark NLP 4.3.0+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[document, token]| +|Output Labels:|[class]| +|Language:|en| +|Size:|466.9 MB| +|Case sensitive:|true| +|Max sentence length:|256| + +## References + +- https://huggingface.co/microsoft/xdoc-base-squad2.0 +- https://arxiv.org/abs/2210.02849 \ No newline at end of file diff --git a/docs/_posts/gadde5300/2023-01-13-bert_embeddings_Italian_Legal_BERT_it.md b/docs/_posts/gadde5300/2023-01-13-bert_embeddings_Italian_Legal_BERT_it.md new file mode 100644 index 00000000000000..1b01dac775c180 --- /dev/null +++ b/docs/_posts/gadde5300/2023-01-13-bert_embeddings_Italian_Legal_BERT_it.md @@ -0,0 +1,74 @@ +--- +layout: model +title: Italian BERT Embedding Cased model +author: John Snow Labs +name: bert_embeddings_Italian_Legal_BERT +date: 2023-01-13 +tags: [it, open_source, embeddings, bert] +task: Embeddings +language: it +edition: Spark NLP 4.2.7 +spark_version: 3.0 +supported: true +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained BERT Embedding model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `Italian-Legal-BERT` is a Italian model originally trained by `dlicari`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/bert_embeddings_Italian_Legal_BERT_it_4.2.7_3.0_1673598434160.zip){:.button.button-orange.button-orange-trans.arr.button-icon} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCol("text") \ + .setOutputCol("document") + +tokenizer = Tokenizer() \ + .setInputCols("document") \ + .setOutputCol("token") + +embeddings = BertEmbeddings.pretrained("bert_embeddings_Italian_Legal_BERT","it") \ + .setInputCols(["document", "token"]) \ + .setOutputCol("embeddings") + +pipeline = Pipeline(stages=[documentAssembler, tokenizer, embeddings]) + +data = spark.createDataFrame([["Adoro Spark NLP"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` + +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|bert_embeddings_Italian_Legal_BERT| +|Compatibility:|Spark NLP 4.2.7+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[sentence]| +|Output Labels:|[bert_sentence]| +|Language:|it| +|Size:|411.6 MB| +|Case sensitive:|true| +|Max sentence length:|128| + +## References + +- https://huggingface.co/dlicari/Italian-Legal-BERT +- https://colab.research.google.com/drive/1aXOmqr70fjm8lYgIoGJMZDsK0QRIL4Lt?usp=sharing \ No newline at end of file diff --git a/docs/_posts/gadde5300/2023-01-13-bert_embeddings_Legal_BERTimbau_base_pt.md b/docs/_posts/gadde5300/2023-01-13-bert_embeddings_Legal_BERTimbau_base_pt.md new file mode 100644 index 00000000000000..085b254003d89e --- /dev/null +++ b/docs/_posts/gadde5300/2023-01-13-bert_embeddings_Legal_BERTimbau_base_pt.md @@ -0,0 +1,74 @@ +--- +layout: model +title: Portuguese BERT Embedding Base Cased model +author: John Snow Labs +name: bert_embeddings_Legal_BERTimbau_base +date: 2023-01-13 +tags: [pt, open_source, embeddings, bert] +task: Embeddings +language: pt +edition: Spark NLP 4.2.7 +spark_version: 3.0 +supported: true +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained BERT Embedding model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `Legal-BERTimbau-base` is a Portuguese model originally trained by `rufimelo`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/bert_embeddings_Legal_BERTimbau_base_pt_4.2.7_3.0_1673598469886.zip){:.button.button-orange.button-orange-trans.arr.button-icon} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCol("text") \ + .setOutputCol("document") + +tokenizer = Tokenizer() \ + .setInputCols("document") \ + .setOutputCol("token") + +embeddings = BertEmbeddings.pretrained("bert_embeddings_Legal_BERTimbau_base","pt") \ + .setInputCols(["document", "token"]) \ + .setOutputCol("embeddings") + +pipeline = Pipeline(stages=[documentAssembler, tokenizer, embeddings]) + +data = spark.createDataFrame([["Eu amo Spark NLP"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` + +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|bert_embeddings_Legal_BERTimbau_base| +|Compatibility:|Spark NLP 4.2.7+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[sentence]| +|Output Labels:|[bert_sentence]| +|Language:|pt| +|Size:|408.6 MB| +|Case sensitive:|true| +|Max sentence length:|128| + +## References + +- https://huggingface.co/rufimelo/Legal-BERTimbau-base +- https://github.com/neuralmind-ai/portuguese-bert/ \ No newline at end of file diff --git a/docs/_posts/gadde5300/2023-01-13-bert_embeddings_Legal_BERTimbau_large_pt.md b/docs/_posts/gadde5300/2023-01-13-bert_embeddings_Legal_BERTimbau_large_pt.md new file mode 100644 index 00000000000000..56027882b3f740 --- /dev/null +++ b/docs/_posts/gadde5300/2023-01-13-bert_embeddings_Legal_BERTimbau_large_pt.md @@ -0,0 +1,74 @@ +--- +layout: model +title: Portuguese BERT Embedding Large Cased model +author: John Snow Labs +name: bert_embeddings_Legal_BERTimbau_large +date: 2023-01-13 +tags: [pt, open_source, embeddings, bert] +task: Embeddings +language: pt +edition: Spark NLP 4.2.7 +spark_version: 3.0 +supported: true +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained BERT Embedding model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `Legal-BERTimbau-large` is a Portuguese model originally trained by `rufimelo`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/bert_embeddings_Legal_BERTimbau_large_pt_4.2.7_3.0_1673598547818.zip){:.button.button-orange.button-orange-trans.arr.button-icon} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCol("text") \ + .setOutputCol("document") + +tokenizer = Tokenizer() \ + .setInputCols("document") \ + .setOutputCol("token") + +embeddings = BertEmbeddings.pretrained("bert_embeddings_Legal_BERTimbau_large","pt") \ + .setInputCols(["document", "token"]) \ + .setOutputCol("embeddings") + +pipeline = Pipeline(stages=[documentAssembler, tokenizer, embeddings]) + +data = spark.createDataFrame([["Eu amo Spark NLP"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` + +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|bert_embeddings_Legal_BERTimbau_large| +|Compatibility:|Spark NLP 4.2.7+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[sentence]| +|Output Labels:|[bert_sentence]| +|Language:|pt| +|Size:|1.3 GB| +|Case sensitive:|true| +|Max sentence length:|128| + +## References + +- https://huggingface.co/rufimelo/Legal-BERTimbau-large +- https://github.com/neuralmind-ai/portuguese-bert/ \ No newline at end of file diff --git a/docs/_posts/gadde5300/2023-01-13-bert_embeddings_bert_large_portuguese_cased_legal_mlm_pt.md b/docs/_posts/gadde5300/2023-01-13-bert_embeddings_bert_large_portuguese_cased_legal_mlm_pt.md new file mode 100644 index 00000000000000..10a05bc37f2f98 --- /dev/null +++ b/docs/_posts/gadde5300/2023-01-13-bert_embeddings_bert_large_portuguese_cased_legal_mlm_pt.md @@ -0,0 +1,79 @@ +--- +layout: model +title: Portuguese BERT Embedding Large Cased model +author: John Snow Labs +name: bert_embeddings_bert_large_portuguese_cased_legal_mlm +date: 2023-01-13 +tags: [pt, open_source, embeddings, bert] +task: Embeddings +language: pt +edition: Spark NLP 4.2.7 +spark_version: 3.0 +supported: true +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained BERT Embedding model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `bert-large-portuguese-cased-legal-mlm` is a Portuguese model originally trained by `stjiris`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/bert_embeddings_bert_large_portuguese_cased_legal_mlm_pt_4.2.7_3.0_1673598653906.zip){:.button.button-orange.button-orange-trans.arr.button-icon} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCol("text") \ + .setOutputCol("document") + +tokenizer = Tokenizer() \ + .setInputCols("document") \ + .setOutputCol("token") + +embeddings = BertEmbeddings.pretrained("bert_embeddings_bert_large_portuguese_cased_legal_mlm","pt") \ + .setInputCols(["document", "token"]) \ + .setOutputCol("embeddings") + +pipeline = Pipeline(stages=[documentAssembler, tokenizer, embeddings]) + +data = spark.createDataFrame([["Eu amo Spark NLP"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` + +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|bert_embeddings_bert_large_portuguese_cased_legal_mlm| +|Compatibility:|Spark NLP 4.2.7+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[sentence]| +|Output Labels:|[bert_sentence]| +|Language:|pt| +|Size:|1.3 GB| +|Case sensitive:|true| +|Max sentence length:|128| + +## References + +- https://huggingface.co/stjiris/bert-large-portuguese-cased-legal-mlm +- https://www.SBERT.net +- https://rufimelo99.github.io/SemanticSearchSystemForSTJ/_static/logo.png +- https://github.com/rufimelo99 +- https://www.inesc-id.pt/projects/PR07005/ +- https://www.inesc-id.pt/wp-content/uploads/2019/06/INESC-ID-logo_01.png +- https://rufimelo99.github.io/SemanticSearchSystemForSTJ/ \ No newline at end of file diff --git a/docs/_posts/gadde5300/2023-01-13-bert_embeddings_legal_bert_base_cased_ptbr_pt.md b/docs/_posts/gadde5300/2023-01-13-bert_embeddings_legal_bert_base_cased_ptbr_pt.md new file mode 100644 index 00000000000000..f2b2908a87bc3d --- /dev/null +++ b/docs/_posts/gadde5300/2023-01-13-bert_embeddings_legal_bert_base_cased_ptbr_pt.md @@ -0,0 +1,74 @@ +--- +layout: model +title: Portuguese BERT Embedding Base Cased model +author: John Snow Labs +name: bert_embeddings_legal_bert_base_cased_ptbr +date: 2023-01-13 +tags: [pt, open_source, embeddings, bert] +task: Embeddings +language: pt +edition: Spark NLP 4.2.7 +spark_version: 3.0 +supported: true +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained BERT Embedding model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `legal-bert-base-cased-ptbr` is a Portuguese model originally trained by `dominguesm`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/bert_embeddings_legal_bert_base_cased_ptbr_pt_4.2.7_3.0_1673598724252.zip){:.button.button-orange.button-orange-trans.arr.button-icon} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCol("text") \ + .setOutputCol("document") + +tokenizer = Tokenizer() \ + .setInputCols("document") \ + .setOutputCol("token") + +embeddings = BertEmbeddings.pretrained("bert_embeddings_legal_bert_base_cased_ptbr","pt") \ + .setInputCols(["document", "token"]) \ + .setOutputCol("embeddings") + +pipeline = Pipeline(stages=[documentAssembler, tokenizer, embeddings]) + +data = spark.createDataFrame([["Eu amo Spark NLP"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` + +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|bert_embeddings_legal_bert_base_cased_ptbr| +|Compatibility:|Spark NLP 4.2.7+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[sentence]| +|Output Labels:|[bert_sentence]| +|Language:|pt| +|Size:|472.7 MB| +|Case sensitive:|true| +|Max sentence length:|128| + +## References + +- https://huggingface.co/dominguesm/legal-bert-base-cased-ptbr +- https://ailab.unb.br/victor/lrec2020 \ No newline at end of file diff --git a/docs/_posts/gadde5300/2023-01-13-camembert_embeddings_Italian_Legal_BERT_SC_it.md b/docs/_posts/gadde5300/2023-01-13-camembert_embeddings_Italian_Legal_BERT_SC_it.md new file mode 100644 index 00000000000000..733245c75acdc0 --- /dev/null +++ b/docs/_posts/gadde5300/2023-01-13-camembert_embeddings_Italian_Legal_BERT_SC_it.md @@ -0,0 +1,73 @@ +--- +layout: model +title: Italian Camembert Embedding Cased model +author: John Snow Labs +name: camembert_embeddings_Italian_Legal_BERT_SC +date: 2023-01-13 +tags: [it, open_source, embeddings, camembert] +task: Embeddings +language: it +edition: Spark NLP 4.2.7 +spark_version: 3.0 +supported: true +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained Camembert Embedding model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `Italian-Legal-BERT-SC` is a Italian model originally trained by `dlicari`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/camembert_embeddings_Italian_Legal_BERT_SC_it_4.2.7_3.0_1673591049304.zip){:.button.button-orange.button-orange-trans.arr.button-icon} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCol("text") \ + .setOutputCol("document") + +tokenizer = Tokenizer() \ + .setInputCols("document") \ + .setOutputCol("token") + +embeddings = CamemBertEmbeddings.pretrained("camembert_embeddings_Italian_Legal_BERT_SC","it") \ + .setInputCols(["document", "token"]) \ + .setOutputCol("embeddings") + +pipeline = Pipeline(stages=[documentAssembler, tokenizer, embeddings]) + +data = spark.createDataFrame([["Adoro Spark NLP"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` + +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|camembert_embeddings_Italian_Legal_BERT_SC| +|Compatibility:|Spark NLP 4.2.7+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[sentence]| +|Output Labels:|[bert_sentence]| +|Language:|it| +|Size:|415.1 MB| +|Case sensitive:|true| +|Max sentence length:|128| + +## References + +- https://huggingface.co/dlicari/Italian-Legal-BERT-SC diff --git a/docs/_posts/gadde5300/2023-01-13-camembert_embeddings_legal_camembert_fr.md b/docs/_posts/gadde5300/2023-01-13-camembert_embeddings_legal_camembert_fr.md new file mode 100644 index 00000000000000..5b88891d036767 --- /dev/null +++ b/docs/_posts/gadde5300/2023-01-13-camembert_embeddings_legal_camembert_fr.md @@ -0,0 +1,75 @@ +--- +layout: model +title: French CamemBERT Embedding Cased model +author: John Snow Labs +name: camembert_embeddings_legal_camembert +date: 2023-01-13 +tags: [fr, open_source, embeddings, camembert] +task: Embeddings +language: fr +edition: Spark NLP 4.2.7 +spark_version: 3.0 +supported: true +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained CamemBERT Embedding model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `legal-camembert` is a French model originally trained by `maastrichtlawtech`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/camembert_embeddings_legal_camembert_fr_4.2.7_3.0_1673596318105.zip){:.button.button-orange.button-orange-trans.arr.button-icon} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCol("text") \ + .setOutputCol("document") + +tokenizer = Tokenizer() \ + .setInputCols("document") \ + .setOutputCol("token") + +embeddings = CamemBertEmbeddings.pretrained("camembert_embeddings_legal_camembert","fr") \ + .setInputCols(["document", "token"]) \ + .setOutputCol("embeddings") + +pipeline = Pipeline(stages=[documentAssembler, tokenizer, embeddings]) + +data = spark.createDataFrame([["J'adore Spark NLP"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` + +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|camembert_embeddings_legal_camembert| +|Compatibility:|Spark NLP 4.2.7+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[sentence]| +|Output Labels:|[bert_sentence]| +|Language:|fr| +|Size:|415.6 MB| +|Case sensitive:|true| +|Max sentence length:|128| + +## References + +- https://huggingface.co/maastrichtlawtech/legal-camembert +- https://antoinelouis.co +- https://www.maastrichtuniversity.nl/about-um/faculties/law/research/law-and-tech-lab \ No newline at end of file diff --git a/docs/_posts/gadde5300/2023-01-13-camembert_embeddings_legal_distilcamembert_fr.md b/docs/_posts/gadde5300/2023-01-13-camembert_embeddings_legal_distilcamembert_fr.md new file mode 100644 index 00000000000000..bc36345b06cb21 --- /dev/null +++ b/docs/_posts/gadde5300/2023-01-13-camembert_embeddings_legal_distilcamembert_fr.md @@ -0,0 +1,75 @@ +--- +layout: model +title: French CamemBERT Embedding Cased model +author: John Snow Labs +name: camembert_embeddings_legal_distilcamembert +date: 2023-01-13 +tags: [fr, open_source, embeddings, camembert] +task: Embeddings +language: fr +edition: Spark NLP 4.2.7 +spark_version: 3.0 +supported: true +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained CamemBERT Embedding model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `legal-distilcamembert` is a French model originally trained by `maastrichtlawtech`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/camembert_embeddings_legal_distilcamembert_fr_4.2.7_3.0_1673596347783.zip){:.button.button-orange.button-orange-trans.arr.button-icon} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCol("text") \ + .setOutputCol("document") + +tokenizer = Tokenizer() \ + .setInputCols("document") \ + .setOutputCol("token") + +embeddings = CamemBertEmbeddings.pretrained("camembert_embeddings_legal_distilcamembert","fr") \ + .setInputCols(["document", "token"]) \ + .setOutputCol("embeddings") + +pipeline = Pipeline(stages=[documentAssembler, tokenizer, embeddings]) + +data = spark.createDataFrame([["J'adore Spark NLP"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` + +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|camembert_embeddings_legal_distilcamembert| +|Compatibility:|Spark NLP 4.2.7+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[sentence]| +|Output Labels:|[bert_sentence]| +|Language:|fr| +|Size:|256.0 MB| +|Case sensitive:|true| +|Max sentence length:|128| + +## References + +- https://huggingface.co/maastrichtlawtech/legal-distilcamembert +- https://antoinelouis.co +- https://www.maastrichtuniversity.nl/about-um/faculties/law/research/law-and-tech-lab \ No newline at end of file diff --git a/docs/_posts/gadde5300/2023-01-13-camembert_embeddings_lsg16k_Italian_Legal_BERT_SC_it.md b/docs/_posts/gadde5300/2023-01-13-camembert_embeddings_lsg16k_Italian_Legal_BERT_SC_it.md new file mode 100644 index 00000000000000..d039e30304c01c --- /dev/null +++ b/docs/_posts/gadde5300/2023-01-13-camembert_embeddings_lsg16k_Italian_Legal_BERT_SC_it.md @@ -0,0 +1,73 @@ +--- +layout: model +title: Italian CamemBERT Embedding Cased model +author: John Snow Labs +name: camembert_embeddings_lsg16k_Italian_Legal_BERT_SC +date: 2023-01-13 +tags: [it, open_source, embeddings, camembert] +task: Embeddings +language: it +edition: Spark NLP 4.2.7 +spark_version: 3.0 +supported: true +article_header: + type: cover +use_language_switcher: "Python-Scala-Java" +--- + +## Description + +Pretrained CamemBERT Embedding model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. `lsg16k-Italian-Legal-BERT-SC` is a Italian model originally trained by `dlicari`. + +{:.btn-box} + + +[Download](https://s3.amazonaws.com/auxdata.johnsnowlabs.com/public/models/camembert_embeddings_lsg16k_Italian_Legal_BERT_SC_it_4.2.7_3.0_1673597331922.zip){:.button.button-orange.button-orange-trans.arr.button-icon} + +## How to use + + + +
+{% include programmingLanguageSelectScalaPythonNLU.html %} +```python +documentAssembler = DocumentAssembler() \ + .setInputCol("text") \ + .setOutputCol("document") + +tokenizer = Tokenizer() \ + .setInputCols("document") \ + .setOutputCol("token") + +embeddings = CamemBertEmbeddings.pretrained("camembert_embeddings_lsg16k_Italian_Legal_BERT_SC","it") \ + .setInputCols(["document", "token"]) \ + .setOutputCol("embeddings") + +pipeline = Pipeline(stages=[documentAssembler, tokenizer, embeddings]) + +data = spark.createDataFrame([["Adoro Spark NLP"]]).toDF("text") + +result = pipeline.fit(data).transform(data) +``` + +
+ +{:.model-param} +## Model Information + +{:.table-model} +|---|---| +|Model Name:|camembert_embeddings_lsg16k_Italian_Legal_BERT_SC| +|Compatibility:|Spark NLP 4.2.7+| +|License:|Open Source| +|Edition:|Official| +|Input Labels:|[sentence]| +|Output Labels:|[bert_sentence]| +|Language:|it| +|Size:|460.7 MB| +|Case sensitive:|true| +|Max sentence length:|128| + +## References + +- https://huggingface.co/dlicari/lsg16k-Italian-Legal-BERT-SC \ No newline at end of file