/pymc3

Probabilistic Programming in Python. Uses Theano as a backend, supports NUTS and ADVI.

Primary LanguagePythonOtherNOASSERTION

PyMC3

Gitter

Build Status

PyMC3 is a python module for Bayesian statistical modeling and model fitting which focuses on advanced Markov chain Monte Carlo fitting algorithms. Its flexibility and extensibility make it applicable to a large suite of problems.

Check out the getting started guide!

Features

  • Intuitive model specification syntax, for example, x ~ N(0,1) translates to x = Normal(0,1)
  • Powerful sampling algorithms, such as the No U-Turn Sampler, allow complex models with thousands of parameters with little specialized knowledge of fitting algorithms.
  • Variational inference: ADVI for fast approximate posterior estimation as well as mini-batch ADVI for large data sets.
  • Easy optimization for finding the maximum a posteriori (MAP) point
  • Theano features
  • Numpy broadcasting and advanced indexing
  • Linear algebra operators
  • Computation optimization and dynamic C compilation
  • Simple extensibility
  • Transparent support for missing value imputation

Getting started

Installation

The latest release of PyMC3 can be installed from PyPI using pip:

pip install pymc3

Note: Running pip install pymc will install PyMC 2.3, not PyMC3, from PyPI.

The current development branch of PyMC3 can be installed from GitHub, also using pip:

pip install git+https://github.com/pymc-devs/pymc3

To ensure the development branch of Theano is installed alongside PyMC3 (recommended), you can install PyMC3 using the requirements.txt file. This requires cloning the repository to your computer:

git clone https://github.com/pymc-devs/pymc3
cd pymc3
pip install -r requirements.txt

However, if a recent version of Theano has already been installed on your system, you can install PyMC3 directly from GitHub.

Another option is to clone the repository and install PyMC3 using python setup.py install or python setup.py develop.

Dependencies

PyMC3 is tested on Python 2.7 and 3.4 and depends on Theano, NumPy, SciPy, Pandas, and Matplotlib (see requirements.txt for version information).

Optional

In addtion to the above dependencies, the GLM submodule relies on Patsy[http://patsy.readthedocs.io/en/latest/].

`scikits.sparse <https://github.com/njsmith/scikits-sparse>`__ enables sparse scaling matrices which are useful for large problems. Installation on Ubuntu is easy:

sudo apt-get install libsuitesparse-dev
pip install git+https://github.com/njsmith/scikits-sparse.git

On Mac OS X you can install libsuitesparse 4.2.1 via homebrew (see http://brew.sh/ to install homebrew), manually add a link so the include files are where scikits-sparse expects them, and then install scikits-sparse:

brew install suite-sparse
ln -s /usr/local/Cellar/suite-sparse/4.2.1/include/ /usr/local/include/suitesparse
pip install git+https://github.com/njsmith/scikits-sparse.git

Citing PyMC3

Salvatier J, Wiecki TV, Fonnesbeck C. (2016) Probabilistic programming in Python using PyMC3. PeerJ Computer Science 2:e55 https://doi.org/10.7717/peerj-cs.55

Coyle P. (2016) Probabilistic programming and PyMC3. European Scientific Python Conference 2015 (Cambridge, UK) http://adsabs.harvard.edu/abs/2016arXiv160700379C

License

Apache License, Version 2.0

Contributors

See the GitHub contributor page