Skip to main content

Magics for defining stan code in notebooks.

Project description

# jupyterstan

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

The package is heavily based on Arvinds-ds
[stanmagic](https://github.com/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.

## Installation

To install the library:

```
pip install git+https://github.com/janfreyberg/jupyterstan.git
```

To enable the syntax highlighting:

```
jupyter nbextension install --py stan_syntax --sys-prefix
jupyter nbextension enable stan_syntax --py --sys-prefix
```

## Usage

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);
}

```

Then, to use your defined model:

```
fit = paris_female_births.sampling(
data={'male': 251527, 'female': 241945},
iter=1000,
chains=4
)
```


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-0.0.1.tar.gz (11.4 kB view hashes)

Uploaded Source

Built Distribution

jupyterstan-0.0.1-py2.py3-none-any.whl (13.1 kB view hashes)

Uploaded Python 2 Python 3

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