https://github.com/quantco/metalearners

MetaLearners for CATE estimation
https://github.com/quantco/metalearners

Last synced: about 1 year ago
JSON representation

MetaLearners for CATE estimation

• Host: GitHub
• URL: https://github.com/quantco/metalearners
• Owner: Quantco
• License: bsd-3-clause
• Created: 2024-06-10T20:15:58.000Z (almost 2 years ago)
• Default Branch: main
• Last Pushed: 2025-03-08T23:00:28.000Z (about 1 year ago)
• Last Synced: 2025-04-02T09:08:24.156Z (about 1 year ago)
• Language: Python
• Homepage: https://metalearners.readthedocs.io/en/latest/
• Size: 1.99 MB
• Stars: 39
• Watchers: 11
• Forks: 4
• Open Issues: 6
• Metadata Files:
  • Readme: README.md
  • Changelog: CHANGELOG.md
  • License: LICENSE
  • Codeowners: .github/CODEOWNERS

Awesome Lists containing this project

README

          
# metalearners

[![CI](https://github.com/Quantco/metalearners/actions/workflows/ci.yml/badge.svg)](https://github.com/Quantco/metalearners/actions/workflows/ci.yml)

[![Documentation Status](https://readthedocs.org/projects/metalearners/badge/?version=latest)](https://metalearners.readthedocs.io/en/latest/?badge=latest)

[![Conda-forge](https://img.shields.io/conda/vn/conda-forge/metalearners?logoColor=white&logo=conda-forge)](https://anaconda.org/conda-forge/metalearners)

[![PypiVersion](https://img.shields.io/pypi/v/metalearners.svg?logo=pypi&logoColor=white)](https://pypi.org/project/metalearners)

[![codecov.io](https://codecov.io/github/QuantCo/metalearners/coverage.svg?branch=main)](https://codecov.io/github/QuantCo/metalearners?branch=main)

MetaLearners for Conditional Average Treatment Effect (CATE) estimation

The library focuses on providing

- Methodologically sound cross-fitting

- Convenient access to and reuse of base models

- Consistent APIs across Metalearners

- Support for more than binary treatment variants

- Integrations with `pandas`, `shap`, `lime`, `optuna` and soon `onnx`

## Example

```python

df = ...

from metalearners import RLearner

from lightgbm import LGBMClassifier, LGBMRegressor

rlearner = RLearner(

    nuisance_model_factory=LGBMRegressor,

    propensity_model_factory=LGBMClassifier,

    treatment_model_factory=LGBMRegressor,

    is_classification=False,

    n_variants=2,

)

features = ["age", "weight", "height"]

rlearner.fit(df[features], df["outcomes"], df["treatment"])

cate_estimates = rlearner.predict(df[features], is_oos=False)

```

Please refer to our

[docs](https://metalearners.readthedocs.io/en/latest/?badge=latest)

for many more in-depth and reproducible examples.

## Installation

`metalearners` can either be installed via PyPI with

```bash

$ pip install metalearners

```

or via conda-forge with

```bash

$ conda install metalearners -c conda-forge

```

## Development

Development instructions can be found [here](https://metalearners.readthedocs.io/en/latest/development.html).