dynsight.analysis.compute_kl_entropy_multi¶

dynsight.analysis.compute_kl_entropy_multi(data, n_neigh=1, units='bit')[source]¶

Estimate Shannon differential entropy using Kozachenko-Leonenko.

This function works for multivariate distribution. The Kozachenko-Leonenko k-nearest neighbors method approximates differential entropy based on distances to nearest neighbors in the sample space. It’s main advantage is being parameter-free.

Deprecated since version v2025.08.27: This function is deprecated and will be removed after June 2026. Use analysis.shannon() instead.

Parameters:

data (NDArray[np.float64]) – The dataset for which the entropy is to be computed. Shape (n_data, n_dims)
n_neigh (int) – The number of neighbors considered in the KL estimator.
units (Literal['bit', 'nat']) – The units of measure of the output entropy. If “bit”, it is computed with log base 2, if “nat” with natural log.

Returns:

The Shannon differential entropy of the dataset, in bits.

Return type:

float

Example

import numpy as np
from dynsight.analysis import compute_kl_entropy_multi

np.random.seed(1234)
data = np.random.rand(10000, 2)

data_entropy = compute_kl_entropy_multi(data)