Skip to main content

Compute Statistical Measures of Divergence

Project description

Divergence

Divergence is a Python package to compute statistical measures of entropy and divergence from probability distributions and samples.

The following functionality is provided:

  • (Information) Entropy
  • Cross Entropy
  • Relative Entropy or Kullback-Leibler Divergence
  • Jensen-Shannon Divergence

The units in which these entropy and divergence measures are calculated can be specified by the user. This is achieved by the argument log_fun, which accepts a function that calculates the logarithm with respect to a particular base. The following units can be realized by the corresponding choice of the argument log_fun in the entropy and divergence calculation functions:

  • bits: base 2 via np.log2
  • nats: base e via np.log
  • dits: base 10 via np.log10

In a Bayesian context, relative entropy can be used as a measure of the information gained by moving from a prior distribution q to a posterior distribution p.

Installation

    pip install divergence

Examples

See the Jupyter notebook Divergence.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

divergence-0.3.2.tar.gz (9.9 kB view hashes)

Uploaded Source

Built Distribution

divergence-0.3.2-py3-none-any.whl (11.1 kB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page