RRLClassifier

class scikit_weak.classification.RRLClassifier(estimator=ExtraTreeClassifier(), n_estimators=100, resample=False, random_state=None)

A class to perform classification for weakly supervised data, based on the RRL algorithm [1]. The y input to the fit method should be given as an iterable of DiscreteWeakLabel

Parameters

  • estimator (estimator class, default=ExtraTreeClassifier) – Base estimator objects to be fitted. Should support predict and predict_proba

  • n_estimators (int, default=100) – The number of trees to be fitted

  • resample (bool, default=False) – Whether to perform bootstrapping or not

  • random_state (int, default=None) – Random seed

Variables

  • classifiers (list of estimators) – The collection of fitted estimators

  • __ys (list of ndarrays) – The collection of sampled target labels. Each ndarray in ys has the same shape as y

  • __Xs (list of ndarrays) – The collection of bootstrapped datasets. Each ndarray in Xs has the same shape as X. If resample=False, then Xs is empty.

  • __n_classes (int) – The number of unique classes in y

  • __classes (list of int) – The unique classes in y

fit(X, y)

Fit the RRLClassifier model

predict(X)

Returns predictions for the given X

predict_proba(X)

Returns probability distributions for the given X

References

[1] Campagner, A., Ciucci, D., Svensson, C. M., Figge, M. T., & Cabitza, F. (2021).

Ground truthing from multi-rater labeling with three-way decision and possibility theory. Information Sciences, 545, 771-790. https://doi.org/10.1016/j.ins.2020.09.049