Metadata-Version: 2.4
Name: wsp-cheval
Version: 0.3.0
Summary: High-performance discrete-choice (logit) travel demand model evaluation
Author-email: Brian Cheung <brian.cheung@wsp.com>, Ashlyn Clarry <ashlyn.clarry@wsp.com>
License-Expression: BSD-3-Clause
Project-URL: Homepage, https://github.com/wsp-sag/cheval
Project-URL: Issues, https://github.com/wsp-sag/cheval/issues
Classifier: Development Status :: 4 - Beta
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python :: 3
Requires-Python: <3.11,>=3.7
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: pandas<1.5,>=0.24
Requires-Dist: numpy<2,>=1.20
Requires-Dist: astor
Requires-Dist: numba>=0.48
Requires-Dist: numexpr
Requires-Dist: attrs>=19.3
Dynamic: license-file

# Cheval (wsp-cheval)

Cheval is a Python package for high-performance evaluation of discrete-choice (logit) models. It's largely built upon the Pandas, NumPy, and NumExpr packages; along with some custom Numba code for performance-critical bottlenecks.

The name is an acronym for "CHoice EVALuator" but has a double-meaning as _cheval_ is the French word for "horse" - and this package has a lot of horsepower! It has been designed for use in travel demand modelling, specifically microsimulated discrete choice models that need to process hundreds of thousands of records through a logit model. It also supports "stochastic" models, where the probabilities are the key outputs.

> [!IMPORTANT]
> As of v0.3, this package is imported using `wsp_cheval` instead of `cheval`

## Key features

Cheval contains two main components:

- `cheval.ChoiceModel` which is the main entry point for discrete choice modelling
- `cheval.LinkedDataFrame` which helps to simplify complex utility calculations.

These components can be used together or separately.

Cheval is compatible with Python 3.7+

## Installation

Cheval can be installed with the following command:

``batch
pip install wsp-cheval
```

### With `pip` directly from GitHub

Cheval can be installed directly from GitHub using `pip` by running the following command:

```batch
pip install git+https://github.com/wsp-sag/wsp-cheval.git
```
