Metadata-Version: 2.1
Name: cicriterion
Version: 1.0
Summary: Implementation of the cross-entropy information criterion and related algorithms.
Home-page: https://github.com/nick-terry/CIC
Author: Nick Terry, Youngjun Choe
Author-email: pnterry@uw.edu
License: UNKNOWN
Project-URL: Bug Tracker, https://github.com/nick-terry/CIC/issues
Platform: UNKNOWN
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: GNU General Public License v3 (GPLv3)
Classifier: Operating System :: OS Independent
Requires-Python: >=3.6
Description-Content-Type: text/markdown
Requires-Dist: cvxpy (>=1.1.12)
Requires-Dist: dccp (>=1.0.3)
Requires-Dist: matplotlib (>=3.3.1)
Requires-Dist: numpy (>=1.19.2)
Requires-Dist: pandas (>=1.1.2)
Requires-Dist: scipy (>=1.6.3)

# CIC (Cross-Entropy Information Criterion)

The CIC is a statistic which can be used to design algorithms for the <a href="https://en.wikipedia.org/wiki/Boltzmann_distribution"> Boltzmann approximation problem </a>. In <a href="https://arxiv.org/abs/1704.04315"> our paper </a> the CIC is theoretically developed and an algorithm is provided for using the CIC to solve Boltzmann approximation problems.

<div><img src="https://i.imgur.com/dYqp8oF.png" alt="Successive approximation of a probability distribution using our algorithm."/></div>

This package contains an implementation for an algorithm described in the paper which solves the Boltzmann approximation problem using the CIC.

# Installation and Use

To install the latest CIC implementation using pip, run the following command:

```
pip install cicriterion
```

To see how to use the CIC implementation, you can refer to example scripts in the <a href="https://github.com/nick-terry/CIC"> project's Github repo </a>. The dependencies for the CIC implementation included with the package will cover any dependencies for experiment scripts found on the Github repo as well.


