Howdy all! I'm pleased to announced the release of pomegranate v0.9.0. The focus of this release is on missing value support across all model fitting / structure learning / inference methods and models. This enables you to do everything from fitting a multivariate Gaussian distribution to an incomplete data set (using a GPU if desired!) to learning the structure of a Bayesian network on an incomplete data set, to running Viterbi decoding using a hidden Markov model on a sequence with some missing values.
Read more about it here: http://bit.ly/2CyrXtX Thanks! Jacob
_______________________________________________ scikit-learn mailing list scikit-learn@python.org https://mail.python.org/mailman/listinfo/scikit-learn