Metadata-Version: 2.1
Name: ennemi
Version: 1.0.0a2
Summary: Easy-to-use Nearest Neighbor Estimation of Mutual Information
Home-page: https://polsys.github.io/ennemi/
Author: Petri Laarne
Author-email: petri.laarne@helsinki.fi
License: MIT
Project-URL: Documentation, https://polsys.github.io/ennemi/
Project-URL: Source, https://github.com/polsys/ennemi/
Project-URL: Issues, https://github.com/polsys/ennemi/issues
Project-URL: Zenodo, https://doi.org/10.5281/zenodo.3834018
Keywords: information-theory entropy mutual-information data-analysis scientific
Platform: UNKNOWN
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Science/Research
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.6
Classifier: Programming Language :: Python :: 3.7
Classifier: Programming Language :: Python :: 3.8
Classifier: Topic :: Scientific/Engineering
Classifier: Topic :: Scientific/Engineering :: Information Analysis
Classifier: Topic :: Scientific/Engineering :: Mathematics
Classifier: Topic :: Scientific/Engineering :: Physics
Classifier: Typing :: Typed
Requires-Python: ~=3.6
Description-Content-Type: text/markdown
Requires-Dist: numpy (>=1.17.5)
Requires-Dist: numpy (<2.0)
Provides-Extra: dev
Requires-Dist: scipy (~=1.4) ; extra == 'dev'
Requires-Dist: pandas (~=1.0) ; extra == 'dev'
Requires-Dist: pytest (~=5.4) ; extra == 'dev'
Requires-Dist: mypy (~=0.770) ; extra == 'dev'

This package implements the estimation of mutual information (MI) between
two continuous variables using a nearest-neighbor algorithm
([Kraskov et al. 2004](https://dx.doi.org/10.1103/PhysRevE.69.066138)).
Mutual information is an information-theoretical measure of
dependency between two variables.

The interface is minimal and aimed at practical data analysis:

- Support for masking and time lags between variables
- Conditional MI with arbitrary-dimensional conditioning variables
- Normalization of MI to correlation coefficient scale
- Optional integration with `pandas` data frame types (no install-time dependency)
- Optimized algorithm and parallel processing of multiple estimation tasks

This package depends only on NumPy.
Support for Python 3.6+ on the latest macOS, Ubuntu and Windows versions
is officially tested.

This project is still in **alpha** status and interface changes are possible.
For more information on theoretical background and usage, please see the
[documentation](https://polsys.github.io/ennemi).
If you encounter any problems or have suggestions, please file an issue!

---

This package has been developed at
[Institute for Atmospheric and Earth System Research (INAR)](https://www.helsinki.fi/en/inar-institute-for-atmospheric-and-earth-system-research),
University of Helsinki.


