Expectation Maximization Python Packages

pycpd

Pure Numpy Implementation of the Coherent Point Drift Algorithm

awesomepmc

Copula-based pairwise Markov chains for non-Gaussian time-series classification, estimation, imputation and forecasting.

awesomepkf

Python library for pairwise Kalman filtering and smoothing on Gaussian pairwise Markov models: linear PKF, nonlinear EPKF/UPKF/PPF, six equivalent linear smoothers, EM learning and back-action tests. Reference code for two companion papers.

ppca-py

[ppca-cpp] Probabilistic PCA (PPCA) with missing-data support — fast C++ core, clean Python API.

biocpd

Incorporating statistical shape models into the Coherent Point Drift algorithm

tgmm

tgmm: A Gaussian Mixture Model based on Expectation-Maximisation implemented with PyTorch

unsupervised-multimodal-trajectory-modeling

We use EM for a mixture of state space models to perform unsupervised clustering of short trajectories.