Scikit-learn Classifier

This module is a wrapper of scikit-learn library. You can initialize classifier instance as same as when initialize scikit-learn instance. Initialize arguments of scikit-learn is fully accept.

class sentivi.classifier.sklearn_clf.ScikitLearnClassifier(num_labels: int = 3, *args, **kwargs)

Scikit-Learn Classifier-based

__init__(num_labels: int = 3, *args, **kwargs)

Initialize ScikitLearnClassifier instance

Parameters

  • num_labels – number of polarities

  • args – arbitrary arguments

  • kwargs – arbitrary keyword arguments

forward(data, *args, **kwargs)

Train and evaluate ScikitLearnClassifier instance

Parameters

  • data – Output of TextEncoder

  • args – arbitrary arguments

  • kwargs – arbitrary keyword arguments

Returns

Training and evaluating result

Return type

str

load(model_path, *args, **kwargs)

Load model from disk

Parameters

  • model_path – path to pre-trained model path

  • args – arbitrary arguments

  • kwargs – arbitrary keyword arguments

Returns

predict(x, *args, **kwargs)

Predict polarities given sentences

Parameters

  • x – TextEncoder.predict output

  • args – arbitrary arguments

  • kwargs – arbitrary keyword arguments

Returns

list of polarities

Return type

list

save(save_path, *args, **kwargs)

Save model to disk

Parameters

  • save_path – path to save model

  • args – arbitrary arguments

  • kwargs – arbitrary keyword arguments

Returns