Friendly modelling API

PyMC3 allows you to write down models using an intuitive syntax to describe a data generating process.

Cutting edge algorithms and model building blocks

Fit your model using gradient-based MCMC algorithms like NUTS, using ADVI for fast approximate inference — including minibatch-ADVI for scaling to large datasets — or using Gaussian processes to build Bayesian nonparametric models.

import pymc3 as pm

X, y = linear_training_data()
with pm.Model() as linear_model:
    weights = pm.Normal('weights', mu=0, sigma=1)
    noise = pm.Gamma('noise', alpha=2, beta=1)
    y_observed = pm.Normal('y_observed',
                mu=X.dot(weights),
                sigma=noise,
                observed=y)

    prior = pm.sample_prior_predictive()
    posterior = pm.sample()
    posterior_pred = pm.sample_posterior_predictive(posterior)

Installation

Via conda-forge:

conda install -c conda-forge pymc3

Via pypi:

pip install pymc3

Latest (unstable):

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

In-Depth Guides

Probability Distributions

PyMC3 includes a comprehensive set of pre-defined statistical distributions that can be used as model building blocks.

Gaussian Processes

Sometimes an unknown parameter or variable in a model is not a scalar value or a fixed-length vector, but a function. A Gaussian process (GP) can be used as a prior probability distribution whose support is over the space of continuous functions. PyMC3 provides rich support for defining and using GPs.

Variational Inference

Variational inference saves computational cost by turning a problem of integration into one of optimization. PyMC3's variational API supports a number of cutting edge algorithms, as well as minibatch for scaling to large datasets.

PyMC3 and Theano

Theano is the deep-learning library PyMC3 uses to construct probability distributions and then access the gradient in order to implement cutting edge inference algorithms. More advanced models may be built by understanding this layer.

License

PyMC3 is licensed under the Apache License, V2.

Citing PyMC3

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

See Google Scholar for a continuously updated list of papers citing PyMC3.

Support and sponsors

PyMC3 is a non-profit project under NumFOCUS umbrella. If you want to support PyMC3 financially, you can donate here.