sitmo

4 ms·

1. 1.▲
   by sitmo 1y ago

   A quick note on how cgmm relates to existing tools: * scikit-learn's GaussianMixture models the unconditional distribution of data. cgmm, on the other hand, models conditional distributions (p(y|x)), which makes it more suitable for re
2. 2.▲
   Show HN: OS Library for Conditional Gaussian Mixture Modelling in Python

   2 points by sitmo 1y ago|1 comments
3. 3.▲
   by sitmo 13y ago

   Works nice. I got a ping of 1ms for my server, so you must be in Amsterdam / Haarlem? Or is there some multi region distributed magic going on?