Skip to main content

Magics for defining stan code in notebooks.

Project description


jupyterstan is a package to help development of Stan models (using pystan) in jupyter notebooks.

The package is heavily based on Arvinds-ds stanmagic package, but provides an interface that simply returns a pystan.Model object.

In addition, it bundles Arvinds-ds stan_code_helper package to improve syntax highlighting for stan cells.


  • Stan language syntax highlighting in all cells beginning with %%stan
  • Compile a stan model and save it as a pystan variable by running a %%stan cell
  • No longer worry about model_code, reading in stan files, etc.


To install the library:

pip install jupyterstan


To use the magic in your notebook, you need to lead the extension:

%load_ext jupyterstan

To define a stan model inside a jupyter notebook, start a cell with the %%stan magic. You can also provide a variable name, which is the variable name that the pystan.Model object will be assigned to. For example:

%%stan paris_female_births
data {
    int male;
    int female;

parameters {
    real<lower=0, upper=1> p;

model {
    female ~ binomial(male + female, p);

When you run this cell, jupyterstan will create a pystan Model object, which will compile your model and allow you to sample from it. To use your compiled model:

fit = paris_female_births.sampling(
    data={'male': 251527, 'female': 241945},

Project details

Download files

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

Source Distribution

jupyterstan-0.2.0.tar.gz (11.6 kB view hashes)

Uploaded source

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page