Source code for text_classification.featurizer.base

from abc import ABC, abstractmethod

import dill


class BaseFeaturizer(ABC):
    """
    Base class that all featurizer classes should inherit from to ensure
    uniformity.
    """

    COARSE_POS_TAGS = ['ADJ', 'ADP', 'ADV', 'AUX', 'CONJ', 'CCONJ', 'DET',
                       'INTJ', 'NOUN', 'NUM', 'PART', 'PRON', 'PROPN',
                       'PUNCT', 'SCONJ', 'SYM', 'VERB', 'X', 'SPACE']

    @abstractmethod
    def extract_features(self, preprocessor, exclude=set()):
        """
        Should add two fields to the preprocessor's instances:
        :code:`feature_vector` and :code:`feature_names`, where
        :code:`feature_vector` is a list of numerical values
        (=feature values) and :code:`feature_names` is a list of
        strings (=brief description of features).

        :param preprocessor: Preprocessor containing samples to featurize.
        :type preprocessor: BasePreprocessor
        :param exclude: Set of features that should be excluded from
            resulting feature vectors.
        :param exclude: Set[str]
        """
        pass

    @abstractmethod
    def add_feature(self, feature_extraction_function):
        """
        Should add a custom feature extraction function to the ones
        defined in the inheriting class.

        :param feature_extraction_function: Custom function that
            extracts features from text.
        :type feature_extraction_function: function
        """
        pass

    def save(self, filename):
        """
        Saves current featurizer instance in binary format.

        :param filename: Name of the file where the featurizer should be
            saved.
        :type filename: str
        """
        with open(filename, "wb") as file:
            dill.dump(self, file)

    @classmethod
    def load(cls, filename):
        """
        Loads a previously saved featurizer from a binary file.

        :param filename: Name of the binary file that the featurizer
            should be loaded from.
        :type filename: str

        :return: Classifier instance.
        """
        with open(filename, "rb") as file:
            featurizer = dill.load(file)

        return featurizer