7.3.4. mclearn.heuristics.qbb_margin_h

mclearn.heuristics.qbb_margin_h(X_training_candidates, **kwargs)[source]

Return the candidate with the smallest average margin.

We first use bagging to train k classifiers. The margin is then defined as the average difference between the two largest values in the prediction vector.

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