[](https://doi.org/10.5281/zenodo.3834019)
This is the first **alpha** release of `ennemi`: _easy-to-use nearest neighbor estimation of mutual information_. Mutual information (MI) can be used to find non-linear correlations between variables, and this Python 3 package is designed to fit into your data analysis workflow.
Features
- Mutual information between one or more variable pairs
- Conditional mutual information, with as many conditioning variables as you like
- Time lags between variables
- Masking of observations, integrated with time lags
- Normalization of results to a correlation coefficient scale
- Integration with `pandas` data frames and series (optional, no runtime dependency)
- Optimized and parallelized algorithm, written in pure Python with `NumPy`
Installation
This package is available on PyPI. To install it, execute
pip install ennemi
on your Python installation. Python 3.6 and later on Linux, macOS and Windows are supported.
Work in progress
This package is still in development. While the algorithm itself is reliable, the interface may still change between alpha versions. If you encounter problems or have suggestions, please file an issue!