pymc 5.27.1

Probabilistic Programming in Python: Bayesian Modeling and Probabilistic Machine Learning with PyTensor

Stars: 9480, Watchers: 9480, Forks: 2208, Open Issues: 461

The pymc-devs/pymc repo was created 16 years ago and the last code push was Yesterday.
The project is extremely popular with a mindblowing 9480 github stars!

How to Install pymc

You can install pymc using pip

pip install pymc

or add it to a project with poetry

poetry add pymc

Package Details

Author

None

License

Apache License, Version 2.0

Homepage

http://github.com/pymc-devs/pymc

PyPi:

https://pypi.org/project/pymc/

GitHub Repo:

https://github.com/pymc-devs/pymc

Classifiers

• Scientific/Engineering
• Scientific/Engineering/Mathematics

Related Packages

Errors

A list of common pymc errors.

Code Examples

Here are some pymc code examples and snippets.

GitHub Issues

The pymc package has 461 open issues on GitHub

• Derive Categorical from Argmin/Argmax
• Fix multi-output IfElse logp for non-lazy backends
• Improve links and references in dims module notebook
• Fix model_to_graphviz for Flat variables
• logprob for non-overlapping switch accepts equivalent spellings
• Add rewrites with RV reparametrization tricks
• ENH: support non-overlapping switch logp transforms with non-zero thresholds (x ? k)
• ENH: general framework for non-overlapping switch logp transforms (monotone branches)
• ENH: support non-overlapping switch logp with scaling on both branches
• ENH: extending equivalent non-overlapping switch logp forms
• Implements more Dims Distributions
• SMC Multiprocessing and Progress Bar Refactor
• Feature/transform aware bounds checking
• Replace legacy seeded_test fixture with NumPy Generator in tests
• Respect value dtype in partial_observed_rv_logprob

See more issues on GitHub