Category Encoders now on conda forge

My scikit-learn compatible library of categorical data encoders (category_encoders) is now published on conda forge!  Conda, if you didn’t know, is an open source package manager for python (and other things) developed primarily by continuum analytics.  Thanks to continuum developer https://github.com/bollwyvl for doing pretty much all of the work to get it working. Check out the […]

Read More

Twitter-Pandas, first release

Thanks to some great help from contributors, we’ve just pushed the first release of twitter pandas, v0.0.1. The first release is aimed at replicating the data-providing (no create/update/delete functions) from the tweepy API with the git-pandas style pandas interface. To install twitterpandas, just use pip pip install twitterpandas And then you can use it right […]

Read More

Mixed-mode estimation in petersburg

A couple of months ago I posted an overview of simple estimation of hierarchical events using python and petersburg.  At the time it probably seemed a little bit trivial, just building a structured frequency model and drawing samples from it. But I have finally implemented the next step to complete the intended functionality.  This post […]

Read More