release notes for ocr 4.3.1

docs/en/spark_ocr_versions/ocr_release_notes.md

@@ -13,6 +13,32 @@ sidebar:
 
 <div class="h3-box" markdown="1">
 
+## 4.3.1
+
+Release date: 17-02-2023
+
+We're glad to announce that Visual NLP 😎 4.3.1 has been released.
+
+### Highlights
+* ImageTextCleaner & ImageTableDetector have improved memory consumption.
+* New Annotators supported in LightPipelines.
+* Table extraction from Digital PDFs pipeline now entirely supported as a LightPipeline.
+
+### ImageTextCleaner & ImageTableDetector improved memory consumption
+* ImageTextCleaner & ImageTableDetector improved memory consumption: we reduced about 30% the memory consumption for this annotator making it more memory friendly and enabling running on memory constrained environments like Colab.
+
+### New Annotators supported in LightPipelines
+Now the following annotators are supported in LightPipelines,
+* PdfToHocr,
+* HocrTokenizer,
+* ImageTableDetector,
+* ImageScaler,
+* HocrToTextTable,
+
+### Table extraction from Digital PDFs pipeline now entirely supported as a LightPipeline.
+* Our Table Extraction from digital PDFs pipeline now supports running as a LightPipeline, check the updated notebook: [SparkOCRPdfToTable.ipynb](https://github.com/JohnSnowLabs/spark-ocr-workshop/blob/master/jupyter/SparkOCRPdfToTable.ipynb)
+
+This release is compatible with Spark NLP for Healthcare 4.3.0, and Spark NLP 4.3.0.
 
 ## 4.3.0
 

docs/en/spark_ocr_versions/release_notes_4_3_1.md

@@ -0,0 +1,49 @@
+---
+layout: docs
+header: true
+seotitle: Visual NLP(Spark OCR)
+title: Visual NLP(Spark OCR) release notes 4.3.1
+permalink: /docs/en/spark_ocr_versions/release_notes_4_3_1
+key: docs-ocr-release-notes
+modify_date: "2023-02-17"
+show_nav: true
+sidebar:
+    nav: spark-ocr
+---
+
+<div class="h3-box" markdown="1">
+
+## 4.3.1
+
+Release date: 17-02-2023
+
+We're glad to announce that Visual NLP 😎 4.3.1 has been released.
+
+### Highlights
+* ImageTextCleaner & ImageTableDetector have improved memory consumption.
+* New Annotators supported in LightPipelines.
+* Table extraction from Digital PDFs pipeline now entirely supported as a LightPipeline.
+
+### ImageTextCleaner & ImageTableDetector improved memory consumption
+* ImageTextCleaner & ImageTableDetector improved memory consumption: we reduced about 30% the memory consumption for this annotator making it more memory friendly and enabling running on memory constrained environments like Colab.
+
+### New Annotators supported in LightPipelines
+Now the following annotators are supported in LightPipelines,
+* PdfToHocr,
+* HocrTokenizer,
+* ImageTableDetector,
+* ImageScaler,
+* HocrToTextTable,
+
+### Table extraction from Digital PDFs pipeline now entirely supported as a LightPipeline.
+* Our Table Extraction from digital PDFs pipeline now supports running as a LightPipeline, check the updated notebook: [SparkOCRPdfToTable.ipynb](https://github.com/JohnSnowLabs/spark-ocr-workshop/blob/master/jupyter/SparkOCRPdfToTable.ipynb)
+
+This release is compatible with Spark NLP for Healthcare 4.3.0, and Spark NLP 4.3.0.
+
+</div><div class="prev_ver h3-box" markdown="1">
+
+## Previous versions
+
+</div>
+
+{%- include docs-sparckocr-pagination.html -%}