Add automatic data preprocessing

polsys / ennemi

Easy Nearest Neighbor Estimation of Mutual Information

https://polsys.github.io/ennemi/

MIT License

35 stars 12 forks source link

Add automatic data preprocessing #54

Closed polsys closed 4 years ago

polsys commented 4 years ago

Implement these two steps that are always done before analysis:

Scale variables to zero mean and unit variance
Add low-amplitude noise

This should be opt-outable by a parameter (preprocess=True), and implemented in both public MI methods. For estimate_entropy, only the noise must be included. For reproducibility, a fixed random seed should be used.

polsys commented 4 years ago

For best results, the scaling should probably be done after masking. With discrete MI, y must obviously not be noised.