Evaluating Shannon entropy/mutual information as a tool for feature selection in Data Science.
Shannon entropy/mutual information as a tool for feature selection in Data Science.
This repo houses, primarily, a blog post, and secondarily, the code required to generate the figures for that blog post.
You should start by reading the blog post in Shannon Entropy.md. That document relies on a couple images stored in the Images subdir.
You can also read this very same blog post, complete with images, @manifolded/hidden-amongst-the-disorder-d48f083a6483?source=friends_link&sk=f9b35c00097eda15cdc34bdc03eb1cfd">here on Medium.
If you came here for the code it is in the file Mutual Information Example.ipynd. This is an iPython notebook, thus it is opened in Jupyter. (In the Python world you must not question this dichotomy!)
I installed my Jupyter whosie-whatsit using Anaconda and usually don’t regret doing so.
This blog post and associated notebook is available under the GNU GPL License v3.0