mixem is a pure-python implementation of the Expectation-Maximization (EM) algorithm for fitting mixtures of probability distributions. It works in Python 2 and Python 3 (tested with 2.7 and 3.5.1) and uses few dependencies (only NumPy and SciPy).

• Easy-to-use and fully-documented API
• Built-in support for several probability distributions
• Easily define custom probability distributions by implementing their probability density function and weighted log-likelihood

Find the mix'EM documentation on ReadTheDocs.

pip install mixem