pxpy 1.0a14

discrete undirected probabilistic models

Copyright (c) 2019 Nico Piatkowski

pxpy

The python library for discrete pairwise undirected graphical models.

FULL-FLEDGED README COMING SOON

Changelog

• 1.0a14: Improved memory management

• 1.0a13: Fixed memory leak with observed data

• 1.0a12: Simplified access to vertex and pairwise marginals

• 1.0a11: Added single variable marginals

• 1.0a10: Optimized library build

• 1.0a9: Enabled sampling with observations; Fixed parameter name in marginal inference

• 1.0a8: Fixed maximum a posteriori estimation; Added custom graph construction

• 1.0a7: Multiple minor fixes; enabled marginal inference with observations; enabled minimal statistics

• 1.0a6: Added manual model creation; enabled training data with missing values

• 1.0a5: Fixed model management

• 1.0a4: Added model access in regularization and proximal hooks

• 1.0a3: Lower GLIBC requirement + removed libgomp dependency (linux)

• 1.0a2: Python 3.5 compatibility

• 1.0a1: Initial release