Skip to main content

Python interface to Stan, a package for Bayesian inference

Project description

Stan logo

pypi version travis-ci build status crate.io

PyStan provides a Python interface to Stan, a package for Bayesian inference using the No-U-Turn sampler, a variant of Hamiltonian Monte Carlo.

For more information on Stan and its modeling language, see the Stan User’s Guide and Reference Manual. PyStan has an interface similar to that of RStan. For an introduction to Stan visit http://mc-stan.org/.

PyStan aims to reproduce the functionality present in RStan. There are a few features present in RStan that have yet to be implemented in PyStan. If you find a feature missing that you use frequently please file a bug report so developers can better direct their efforts.

Similar projects

Installation

NumPy and Cython (version 0.19.1 or greater) are required. matplotlib is optional.

PyStan and the required packages may be installed from the Python Package Index using pip.

pip install numpy Cython
pip install pystan

Alternatively, if Cython (version 0.19 or greater) and NumPy are already available, PyStan may be installed from source with the following commands

git clone https://github.com/stan-dev/pystan.git
cd pystan
python setup.py install

If you encounter an ImportError after compiling from source, try changing out of the source directory before attempting import pystan. For example, on Linux and OS X cd /tmp would work.

Example

import pystan
import numpy as np

schools_code = """
data {
    int<lower=0> J; // number of schools
    real y[J]; // estimated treatment effects
    real<lower=0> sigma[J]; // s.e. of effect estimates
}
parameters {
    real mu;
    real<lower=0> tau;
    real eta[J];
}
transformed parameters {
    real theta[J];
    for (j in 1:J)
        theta[j] <- mu + tau * eta[j];
}
model {
    eta ~ normal(0, 1);
    y ~ normal(theta, sigma);
}
"""

schools_dat = {'J': 8,
               'y': [28,  8, -3,  7, -1,  1, 18, 12],
               'sigma': [15, 10, 16, 11,  9, 11, 10, 18]}

fit = pystan.stan(model_code=schools_code, data=schools_dat,
                  iter=1000, chains=4)

print(fit)

eta = fit.extract(permuted=True)['eta']
np.mean(eta, axis=0)

# if matplotlib is installed (optional, not required), a visual summary and
# traceplot are available
fit.plot()

Project details


Release history Release notifications

History Node

2.17.1.0

History Node

2.17.0.0

History Node

2.16.0.0

History Node

2.15.0.1

History Node

2.15.0.0

History Node

2.14.0.0

History Node

2.12.0.0

History Node

2.11.0.0

History Node

2.10.0.0

History Node

2.9.0.0

History Node

2.8.0.2

History Node

2.8.0.1

History Node

2.8.0.0

History Node

2.7.0.1

History Node

2.7.0.0

History Node

2.6.3.0

History Node

2.6.0.0

History Node

2.5.0.2

History Node

2.5.0.1

History Node

2.5.0.0

History Node

2.4.0.3

History Node

2.4.0.2

History Node

2.4.0.1

History Node

2.4.0.0

History Node

2.3.0.0

History Node

2.2.0.1

History Node

2.2.0.0

History Node

2.1.0.1

History Node

2.1.0.0

History Node

2.0.1.3

History Node

2.0.1.2

This version
History Node

2.0.1.1

History Node

2.0.1.0

History Node

2.0.0.1

History Node

2.0.0.0

History Node

0.2.2

History Node

0.2.1

History Node

0.2.0

History Node

0.1.1

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Filename, size & hash SHA256 hash help File type Python version Upload date
pystan-2.0.1.1.tar.gz (16.9 MB) Copy SHA256 hash SHA256 Source None Nov 19, 2013

Supported by

Elastic Elastic Search Pingdom Pingdom Monitoring Google Google BigQuery Sentry Sentry Error logging CloudAMQP CloudAMQP RabbitMQ AWS AWS Cloud computing DataDog DataDog Monitoring Fastly Fastly CDN DigiCert DigiCert EV certificate StatusPage StatusPage Status page