klGauss

statrl.settings.utils.klGauss(x, y, sig2=1.0)

Kullback-Leibler divergence between two Gaussians of equal variance.

\[\mathrm{kl}(x, y) = \frac{(x - y)^2}{2\sigma^2}\]

Parameters:

• x (float) – Means of the two distributions.

• y (float) – Means of the two distributions.

• sig2 (float, default=1.0) – Common variance \(\sigma^2\).

Returns:

The divergence, in nats.

Return type:

float

See also

klBern, klPoisson, klExp

Examples

>>> klGauss(0.0, 1.0)
0.5