TextClassification.lf

/**
 * @file BertQA.lf
 * @author Vincenzo Barbuto
 * @brief Examples of how to use the NLP library to perform text classification.
 */
target Python

import NLClassifier from "lib/NLP.lf"

/**
 * The `Injector` reactor is responsible for getting text input from the user and passing it to the
 * `NLClassifier` reactor.
 *
 * The `get_text()` method prompts the user to enter text to be classified, and the
 * `reaction(startup)` method sets the `text` output to the user's input.
 *
 * The `ResultPrinter` reactor is responsible for printing the classification results and the
 * inference time. The `reaction(results, inference_time)` method iterates through the
 * classification results and prints the name and score of each result, as well as the inference
 * time.
 *
 * The `main` reactor creates instances of the `NLClassifier`, `Injector`, and `ResultPrinter`
 * reactors, and connects them together to form the text classification pipeline.
 */
reactor Injector {
  output text

  preamble {=
    def get_text(self):
        text = input("Enter text to classify: ")
        return text
  =}

  reaction(startup) -> text {=
    print("Starting TextInput reactor")
    text.set(self.get_text())
  =}
}

reactor ResultPrinter {
  input results
  input inference_time

  reaction(results, inference_time) {=
    for result in results.value:
        print(f"Result: {result['name']}, Score: {result['score']}")
    print(f"Inference time: {inference_time.value} ms")
  =}
}

main reactor {
  classifier = new NLClassifier(
      model = {= os.path.join(os.getcwd(),"models/nlp/classification/bert_classifier.tflite") =})
  injector = new Injector()
  printer = new ResultPrinter()
  injector.text -> classifier.input_data
  classifier.results -> printer.results
  classifier.inference_time -> printer.inference_time
}