Contents

PyMC

PyMC (formerly known as PyMC3) is a probabilistic programming language written in Python. It can be used for Bayesian statistical modeling and probabilistic machine learning.

PyMC performs inference based on advanced Markov chain Monte Carlo and/or variational fitting algorithms.^{[2] [3][4][5][6]} It is a rewrite from scratch of the previous version of the PyMC software.^[7] Unlike PyMC2, which had used Fortran extensions for performing computations, PyMC relies on PyTensor, a Python library that allows defining, optimizing, and efficiently evaluating mathematical expressions involving multi-dimensional arrays. From version 3.8 PyMC relies on ArviZ to handle plotting, diagnostics, and statistical checks. PyMC and Stan are the two most popular probabilistic programming tools.^[8] PyMC is an open source project, developed by the community and has been fiscally sponsored by NumFOCUS.^[9]

PyMC has been used to solve inference problems in several scientific domains, including astronomy,^[10][11] epidemiology,^[12][13] molecular biology,^[14] crystallography,^[15][16] chemistry,^[17] ecology^[18][19] and psychology.^[20] Previous versions of PyMC were also used widely, for example in climate science,^[21] public health,^[22] neuroscience,^[23] and parasitology.^[24][25]

After Theano announced plans to discontinue development in 2017,^[26] the PyMC team evaluated TensorFlow Probability as a computational backend,^[27] but decided in 2020 to fork Theano under the name Aesara.^[28] Large parts of the Theano codebase have been refactored and compilation through JAX^[29] and Numba were added. The PyMC team has released the revised computational backend under the name PyTensor and continues the development of PyMC.^[30]

Inference engines[edit]

PyMC implements non-gradient-based and gradient-based Markov chain Monte Carlo (MCMC) algorithms for Bayesian inference and stochastic, gradient-based variational Bayesian methods for approximate Bayesian inference.

• MCMC-based algorithms:
  • No-U-Turn sampler^[31] (NUTS), a variant of Hamiltonian Monte Carlo and PyMC's default engine for continuous variables
  • Metropolis–Hastings, PyMC's default engine for discrete variables
  • Sequential Monte Carlo for static posteriors
  • Sequential Monte Carlo for approximate Bayesian computation
• Variational inference algorithms:
  • Black-box Variational Inference^[32]

See also[edit]

• Stan is a probabilistic programming language for statistical inference written in C++
• ArviZ a Python library for Exploratory Analysis of Bayesian Models

References[edit]

Further reading[edit]

• Martin, Osvaldo (2024). Bayesian Analysis with Python : A Practical Guide to Probabilistic Modeling (Third ed.). Packt. ISBN 978-1-80512-716-1.

External links[edit]

• PyMC website
• PyMC source, a Git repository hosted on GitHub
• PyTensor is a Python library for defining, optimizing, and efficiently evaluating mathematical expressions involving multi-dimensional arrays.