7.3.5. mclearn.heuristics.qbb_kl_h¶
-
mclearn.heuristics.
qbb_kl_h
(X_training_candidates, **kwargs)[source]¶ Return the candidate with the largest average KL divergence from the mean.
We first use bagging to train k classifiers. We then choose the candidate that has the largest Kullback–Leibler divergence from the average.
Parameters: X_training_candidates (array) – The feature matrix of the potential training candidates. Returns: best_candidate – The index of the best candidate. Return type: int