44bdf9be3b913a1a0e90115e45edbee192d01b90,deslib/des/probabilistic.py,Probabilistic,estimate_competence,#Probabilistic#Any#Any#,113

Before Change

        potential_dists = self.potential_func(dists_organized)
        sum_potential = np.sum(potential_dists)
        competences = np.zeros(self.n_classifiers)

        for clf_index in range(self.n_classifiers):
            // Check if the dynamic frienemy pruning (DFP) should be used used
            if self.DFP_mask[clf_index]:
                temp_competence = np.multiply(self.C_src[:, clf_index], potential_dists)
                competences[clf_index] = np.sum(temp_competence)/sum_potential

        return competences

    def select(self, competences):
        Selects the base classifiers that obtained a competence level higher than the predefined threshold.

After Change

        // TODO: Adapt this function to batch processing
        dists, idx_neighbors = self._get_region_competence(query)
        idx_neighbors = np.atleast_2d(idx_neighbors)
        dists = np.atleast_2d(dists)
        dists_organized = np.take(dists, idx_neighbors)

        //dists_organized = np.array([dists[index] for index in np.argsort(idx_neighbors)])
        potential_dists = self.potential_func(dists_organized)
        sum_potential = np.sum(potential_dists, axis=1)

        competences = np.einsum("ijk,ij->ik", self.C_src[np.newaxis, :, :], potential_dists)
        competences = competences/sum_potential.reshape(-1, 1)

        // competences = np.zeros(self.n_classifiers)
        // for clf_index in range(self.n_classifiers):
        //     // Check if the dynamic frienemy pruning (DFP) should be used used

Instances