docker_images/flair/app/pipelines/token_classification.py

from typing import Any, Dict, List

from app.pipelines import Pipeline
from flair.data import Sentence, Span, Token
from flair.models import SequenceTagger


class TokenClassificationPipeline(Pipeline):
    def __init__(
        self,
        model_id: str,
    ):
        self.tagger = SequenceTagger.load(model_id)

    def __call__(self, inputs: str) -> List[Dict[str, Any]]:
        """
        Args:
            inputs (:obj:`str`):
                a string containing some text
        Return:
            A :obj:`list`:. The object returned should be like [{"entity_group": "XXX", "word": "some word", "start": 3, "end": 6, "score": 0.82}] containing :
                - "entity_group": A string representing what the entity is.
                - "word": A substring of the original string that was detected as an entity.
                - "start": the offset within `input` leading to `answer`. context[start:stop] == word
                - "end": the ending offset within `input` leading to `answer`. context[start:stop] === word
                - "score": A score between 0 and 1 describing how confident the model is for this entity.
        """
        sentence: Sentence = Sentence(inputs)

        self.tagger.predict(sentence)

        entities = []
        for label in sentence.get_labels():
            current_data_point = label.data_point
            if isinstance(current_data_point, Token):
                current_entity = {
                    "entity_group": current_data_point.tag,
                    "word": current_data_point.text,
                    "start": current_data_point.start_position,
                    "end": current_data_point.end_position,
                    "score": current_data_point.score,
                }
                entities.append(current_entity)
            elif isinstance(current_data_point, Span):
                if not current_data_point.tokens:
                    continue
                current_entity = {
                    "entity_group": current_data_point.tag,
                    "word": current_data_point.text,
                    "start": current_data_point.tokens[0].start_position,
                    "end": current_data_point.tokens[-1].end_position,
                    "score": current_data_point.score,
                }
                entities.append(current_entity)

        return entities