kl_divergence: Kullback–Leibler Divergence (KLD)

In nikitagusarov/performancer: Performance Metrics for Discrete Choice Analysis

Kullback–Leibler Divergence (KLD)

Description

Compute Kullback–Leibler Divergence (KLD) using confusion matrix. KL divergence basically just finds the difference between the entropies of the two distributions 'P(y|f)' and 'p(y)'. The inputs are assumed to be expressed in probabilistic terms.

Usage

kl_divergence(y_real, y_predicted)

Arguments

y_real	Observed values (integers) to compare with (in matrix format for multiclass classification).
y_predicted	Predicte values (probabiblities by class).

Value

integer value of Kullback–Leibler Divergence (KLD)

nikitagusarov/performancer documentation built on Jan. 12, 2023, 12:19 a.m.