pusion.core.macro_majority_vote_combiner

class pusion.core.macro_majority_vote_combiner.MacroMajorityVoteCombiner

Bases: pusion.core.combiner.UtilityBasedCombiner

The MacroMajorityVoteCombiner (MAMV) is based on a variation of the general majority vote method. The fusion consists of a decision vector which is given by the majority of the classifiers in the ensemble for a sample. MAMV does not consider outputs for each individual class (macro).

combine(decision_tensor)

Combine decision outputs by majority voting across all classifiers considering the most common classification assignment (macro). Only crisp classification outputs are supported.

Parameters

decision_tensor – numpy.array of shape (n_classifiers, n_samples, n_classes). Tensor of crisp decision outputs by different classifiers per sample.

Returns

A matrix (numpy.array) of crisp class assignments obtained by MAMV. Axis 0 represents samples and axis 1 the class labels which are aligned with axis 2 in decision_tensor input tensor.