PyStan: The Python Interface to Stan

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 at http://mc-stan.org/.

Important links

• HTML documentation: https://pystan.readthedocs.org
• Issue tracker: https://github.com/stan-dev/pystan/issues
• Source code repository: https://github.com/stan-dev/pystan
• Stan: http://mc-stan.org/
• Stan User's Guide and Reference Manual (pdf) available at http://mc-stan.org

Similar projects

• PyMC: http://pymc-devs.github.io/pymc/

Installation

NumPy and Cython (version 0.19 or greater) are required. matplotlib is optional.

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

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()