pygam


Namepygam JSON
Version 0.10.1 PyPI version JSON
download
home_pageNone
SummaryGeneralized Additive Models in Python.
upload_time2025-07-26 17:31:26
maintainerNone
docs_urlNone
authorDaniel Servén Marín, Charlie Brummitt
requires_python<3.14,>=3.9
licenseApache-2.0
keywords
VCS
bugtrack_url
requirements No requirements were recorded.
Travis-CI No Travis.
coveralls test coverage No coveralls.
            

# pyGAM

<a href="https://pygam.readthedocs.io/en/latest/?badge=latest"><img src=imgs/pygam_tensor.png width="250" align="right" /></a>

Generalized Additive Models in Python.

:rocket: **Version 0.10.1 out now!** [See release notes here](https://github.com/dswah/pyGAM/releases).

`pyGAM` is a package for building Generalized Additive Models in Python, with an emphasis on modularity and performance.

The API is designed for users of `scikit-learn` or `scipy`.


|  | **[Documentation](https://pygam.readthedocs.io/en/latest/?badge=latest)** · **[Tutorials](https://pygam.readthedocs.io/en/latest/notebooks/tour_of_pygam.html)** · **[Medium article](https://medium.com/just-another-data-scientist/building-interpretable-models-with-generalized-additive-models-in-python-c4404eaf5515)** |
|---|---|
| **Open&#160;Source** | [![Apache 2.0](https://img.shields.io/badge/license-Apache%202.0-blue.svg)](https://github.com/dswah/pygam/blob/main/LICENSE) [![GC.OS Sponsored](https://img.shields.io/badge/GC.OS-Sponsored%20Project-orange.svg?style=flat&colorA=0eac92&colorB=2077b4)](https://gc-os-ai.github.io/) |
| **Community** | [![!discord](https://img.shields.io/static/v1?logo=discord&label=discord&message=chat&color=lightgreen)](https://discord.gg/Rt8By5Jj) [![!slack](https://img.shields.io/static/v1?logo=linkedin&label=LinkedIn&message=news&color=lightblue)](https://www.linkedin.com/company/german-center-for-open-source-ai) |
| **CI/CD** | [![github-actions](https://img.shields.io/github/actions/workflow/status/dswah/pygam/pypi.yml?logo=github)](https://github.com/dswah/pygam/actions/workflows/pypi.yml) [![readthedocs](https://img.shields.io/readthedocs/pygam?logo=readthedocs)](https://pygam.readthedocs.io/en/latest/?badge=latest) |
| **Code** |  [![!pypi](https://img.shields.io/pypi/v/pygam?color=orange)](https://pypi.org/project/pygam/) [![!conda](https://img.shields.io/conda/vn/conda-forge/pygam)](https://anaconda.org/conda-forge/pygam) [![!python-versions](https://img.shields.io/pypi/pyversions/pygam)](https://www.python.org/) [![!black](https://img.shields.io/badge/code%20style-black-000000.svg)](https://github.com/psf/black)  |
| **Downloads** | ![PyPI - Downloads](https://img.shields.io/pypi/dw/pygam) ![PyPI - Downloads](https://img.shields.io/pypi/dm/pygam) [![Downloads](https://static.pepy.tech/personalized-badge/pygam?period=total&units=international_system&left_color=grey&right_color=blue&left_text=cumulative%20(pypi))](https://pepy.tech/project/pygam) |
| **Citation** | [![!zenodo](https://zenodo.org/badge/DOI/10.5281/zenodo.1208723.svg)](https://doi.org/10.5281/zenodo.1208723) |

## Documentation
- [Official pyGAM Documentation: Read the Docs](https://pygam.readthedocs.io/en/latest/?badge=latest)
- [Building interpretable models with Generalized additive models in Python](https://medium.com/just-another-data-scientist/building-interpretable-models-with-generalized-additive-models-in-python-c4404eaf5515)
<!-----
[pyGAM: Getting started with Generalized Additive Models in Python](https://medium.com/@jpoberhauser/pygam-getting-started-with-generalized-additive-models-in-python-457df5b4705f)
----->

## Installation
```pip install pygam```

### scikit-sparse
To speed up optimization on large models with constraints, it helps to have `scikit-sparse` installed because it contains a slightly faster, sparse version of Cholesky factorization. The import from `scikit-sparse` references `nose`, so you'll need that too.

The easiest way is to use Conda:
```conda install -c conda-forge scikit-sparse nose```

[scikit-sparse project](https://github.com/scikit-sparse/scikit-sparse)

## Contributing - HELP REQUESTED
Contributions are most welcome!

You can help pyGAM in many ways including:

- Working on a [known bug](https://github.com/dswah/pyGAM/labels/bug).
- Trying it out and reporting bugs or what was difficult.
- Helping improve the documentation.
- Writing new [distributions](https://github.com/dswah/pyGAM/blob/master/pygam/distributions.py), and [link functions](https://github.com/dswah/pyGAM/blob/master/pygam/links.py).
- If you need some ideas, please take a look at the [issues](https://github.com/dswah/pyGAM/issues).


To start:
- **fork the project** and cut a new branch
- **install** `pygam`, editable with developer **dependencies** (in a new python environment)

```
pip install --upgrade pip
pip install -e ".[dev]"
```

Make some changes and write a test...
- **Test** your contribution (eg from the `.../pyGAM`):
```py.test -s```
- When you are happy with your changes, make a **pull request** into the `master` branch of the main project.


## About
Generalized Additive Models (GAMs) are smooth semi-parametric models of the form:

![alt tag](http://latex.codecogs.com/svg.latex?g\(\mathbb{E}\[y|X\]\)=\beta_0+f_1(X_1)+f_2(X_2)+\dots+f_p(X_p))

where `X.T = [X_1, X_2, ..., X_p]` are independent variables, `y` is the dependent variable, and `g()` is the link function that relates our predictor variables to the expected value of the dependent variable.

The feature functions `f_i()` are built using **penalized B splines**, which allow us to **automatically model non-linear relationships** without having to manually try out many different transformations on each variable.

<img src=imgs/pygam_basis.png>

GAMs extend generalized linear models by allowing non-linear functions of features while maintaining additivity. Since the model is additive, it is easy to examine the effect of each `X_i` on `Y` individually while holding all other predictors constant.

The result is a very flexible model, where it is easy to incorporate prior knowledge and control overfitting.

## Citing pyGAM
Please consider citing pyGAM if it has helped you in your research or work:

Daniel Servén, & Charlie Brummitt. (2018, March 27). pyGAM: Generalized Additive Models in Python. Zenodo. [DOI: 10.5281/zenodo.1208723](http://doi.org/10.5281/zenodo.1208723)

BibTex:
```
@misc{daniel\_serven\_2018_1208723,
  author       = {Daniel Servén and
                  Charlie Brummitt},
  title        = {pyGAM: Generalized Additive Models in Python},
  month        = mar,
  year         = 2018,
  doi          = {10.5281/zenodo.1208723},
  url          = {https://doi.org/10.5281/zenodo.1208723}
}
```

## References
1. Simon N. Wood, 2006
Generalized Additive Models: an introduction with R

0. Hastie, Tibshirani, Friedman
The Elements of Statistical Learning
https://www.sas.upenn.edu/~fdiebold/NoHesitations/BookAdvanced.pdf

0. James, Witten, Hastie and Tibshirani
An Introduction to Statistical Learning
http://www-bcf.usc.edu/~gareth/ISL/ISLR%20Sixth%20Printing.pdf

0. Paul Eilers & Brian Marx, 1996
Flexible Smoothing with B-splines and Penalties
https://sites.stat.washington.edu/courses/stat527/s14/readings/EilersMarx_StatSci_1996.pdf

0. Kim Larsen, 2015
GAM: The Predictive Modeling Silver Bullet
http://multithreaded.stitchfix.com/assets/files/gam.pdf

0. Deva Ramanan, 2008
UCI Machine Learning: Notes on IRLS
http://www.ics.uci.edu/~dramanan/teaching/ics273a_winter08/homework/irls_notes.pdf

0. Paul Eilers & Brian Marx, 2015
International Biometric Society: A Crash Course on P-splines
https://multithreaded.stitchfix.com/assets/files/gam.pdf

0. Keiding, Niels, 1991
Age-specific incidence and prevalence: a statistical perspective
https://academic.oup.com/jrsssa/article-abstract/154/3/371/7106499


<!---http://www.cs.princeton.edu/courses/archive/fall11/cos323/notes/cos323_f11_lecture09_svd.pdf--->

<!---http://www.stats.uwo.ca/faculty/braun/ss3859/notes/Chapter4/ch4.pdf--->

<!---http://www.stat.berkeley.edu/~census/mlesan.pdf--->

<!---http://web.mit.edu/hyperbook/Patrikalakis-Maekawa-Cho/node17.html---> <!--- this helped me get spline gradients--->

<!---https://scikit-sparse.readthedocs.io/en/latest/overview.html#developers--->

<!---https://vincentarelbundock.github.io/Rdatasets/datasets.html---> <!--- R Datasets!--->

            

Raw data

            {
    "_id": null,
    "home_page": null,
    "name": "pygam",
    "maintainer": null,
    "docs_url": null,
    "requires_python": "<3.14,>=3.9",
    "maintainer_email": null,
    "keywords": null,
    "author": "Daniel Serv\u00e9n Mar\u00edn, Charlie Brummitt",
    "author_email": null,
    "download_url": "https://files.pythonhosted.org/packages/11/62/ef01ba5f1ad7225b47565e06dc52d46901812ce92fd6400a68ced17dd0b2/pygam-0.10.1.tar.gz",
    "platform": null,
    "description": "\n\n# pyGAM\n\n<a href=\"https://pygam.readthedocs.io/en/latest/?badge=latest\"><img src=imgs/pygam_tensor.png width=\"250\" align=\"right\" /></a>\n\nGeneralized Additive Models in Python.\n\n:rocket: **Version 0.10.1 out now!** [See release notes here](https://github.com/dswah/pyGAM/releases).\n\n`pyGAM` is a package for building Generalized Additive Models in Python, with an emphasis on modularity and performance.\n\nThe API is designed for users of `scikit-learn` or `scipy`.\n\n\n|  | **[Documentation](https://pygam.readthedocs.io/en/latest/?badge=latest)** \u00b7 **[Tutorials](https://pygam.readthedocs.io/en/latest/notebooks/tour_of_pygam.html)** \u00b7 **[Medium article](https://medium.com/just-another-data-scientist/building-interpretable-models-with-generalized-additive-models-in-python-c4404eaf5515)** |\n|---|---|\n| **Open&#160;Source** | [![Apache 2.0](https://img.shields.io/badge/license-Apache%202.0-blue.svg)](https://github.com/dswah/pygam/blob/main/LICENSE) [![GC.OS Sponsored](https://img.shields.io/badge/GC.OS-Sponsored%20Project-orange.svg?style=flat&colorA=0eac92&colorB=2077b4)](https://gc-os-ai.github.io/) |\n| **Community** | [![!discord](https://img.shields.io/static/v1?logo=discord&label=discord&message=chat&color=lightgreen)](https://discord.gg/Rt8By5Jj) [![!slack](https://img.shields.io/static/v1?logo=linkedin&label=LinkedIn&message=news&color=lightblue)](https://www.linkedin.com/company/german-center-for-open-source-ai) |\n| **CI/CD** | [![github-actions](https://img.shields.io/github/actions/workflow/status/dswah/pygam/pypi.yml?logo=github)](https://github.com/dswah/pygam/actions/workflows/pypi.yml) [![readthedocs](https://img.shields.io/readthedocs/pygam?logo=readthedocs)](https://pygam.readthedocs.io/en/latest/?badge=latest) |\n| **Code** |  [![!pypi](https://img.shields.io/pypi/v/pygam?color=orange)](https://pypi.org/project/pygam/) [![!conda](https://img.shields.io/conda/vn/conda-forge/pygam)](https://anaconda.org/conda-forge/pygam) [![!python-versions](https://img.shields.io/pypi/pyversions/pygam)](https://www.python.org/) [![!black](https://img.shields.io/badge/code%20style-black-000000.svg)](https://github.com/psf/black)  |\n| **Downloads** | ![PyPI - Downloads](https://img.shields.io/pypi/dw/pygam) ![PyPI - Downloads](https://img.shields.io/pypi/dm/pygam) [![Downloads](https://static.pepy.tech/personalized-badge/pygam?period=total&units=international_system&left_color=grey&right_color=blue&left_text=cumulative%20(pypi))](https://pepy.tech/project/pygam) |\n| **Citation** | [![!zenodo](https://zenodo.org/badge/DOI/10.5281/zenodo.1208723.svg)](https://doi.org/10.5281/zenodo.1208723) |\n\n## Documentation\n- [Official pyGAM Documentation: Read the Docs](https://pygam.readthedocs.io/en/latest/?badge=latest)\n- [Building interpretable models with Generalized additive models in Python](https://medium.com/just-another-data-scientist/building-interpretable-models-with-generalized-additive-models-in-python-c4404eaf5515)\n<!-----\n[pyGAM: Getting started with Generalized Additive Models in Python](https://medium.com/@jpoberhauser/pygam-getting-started-with-generalized-additive-models-in-python-457df5b4705f)\n----->\n\n## Installation\n```pip install pygam```\n\n### scikit-sparse\nTo speed up optimization on large models with constraints, it helps to have `scikit-sparse` installed because it contains a slightly faster, sparse version of Cholesky factorization. The import from `scikit-sparse` references `nose`, so you'll need that too.\n\nThe easiest way is to use Conda:\n```conda install -c conda-forge scikit-sparse nose```\n\n[scikit-sparse project](https://github.com/scikit-sparse/scikit-sparse)\n\n## Contributing - HELP REQUESTED\nContributions are most welcome!\n\nYou can help pyGAM in many ways including:\n\n- Working on a [known bug](https://github.com/dswah/pyGAM/labels/bug).\n- Trying it out and reporting bugs or what was difficult.\n- Helping improve the documentation.\n- Writing new [distributions](https://github.com/dswah/pyGAM/blob/master/pygam/distributions.py), and [link functions](https://github.com/dswah/pyGAM/blob/master/pygam/links.py).\n- If you need some ideas, please take a look at the [issues](https://github.com/dswah/pyGAM/issues).\n\n\nTo start:\n- **fork the project** and cut a new branch\n- **install** `pygam`, editable with developer **dependencies** (in a new python environment)\n\n```\npip install --upgrade pip\npip install -e \".[dev]\"\n```\n\nMake some changes and write a test...\n- **Test** your contribution (eg from the `.../pyGAM`):\n```py.test -s```\n- When you are happy with your changes, make a **pull request** into the `master` branch of the main project.\n\n\n## About\nGeneralized Additive Models (GAMs) are smooth semi-parametric models of the form:\n\n![alt tag](http://latex.codecogs.com/svg.latex?g\\(\\mathbb{E}\\[y|X\\]\\)=\\beta_0+f_1(X_1)+f_2(X_2)+\\dots+f_p(X_p))\n\nwhere `X.T = [X_1, X_2, ..., X_p]` are independent variables, `y` is the dependent variable, and `g()` is the link function that relates our predictor variables to the expected value of the dependent variable.\n\nThe feature functions `f_i()` are built using **penalized B splines**, which allow us to **automatically model non-linear relationships** without having to manually try out many different transformations on each variable.\n\n<img src=imgs/pygam_basis.png>\n\nGAMs extend generalized linear models by allowing non-linear functions of features while maintaining additivity. Since the model is additive, it is easy to examine the effect of each `X_i` on `Y` individually while holding all other predictors constant.\n\nThe result is a very flexible model, where it is easy to incorporate prior knowledge and control overfitting.\n\n## Citing pyGAM\nPlease consider citing pyGAM if it has helped you in your research or work:\n\nDaniel Serv\u00e9n, & Charlie Brummitt. (2018, March 27). pyGAM: Generalized Additive Models in Python. Zenodo. [DOI: 10.5281/zenodo.1208723](http://doi.org/10.5281/zenodo.1208723)\n\nBibTex:\n```\n@misc{daniel\\_serven\\_2018_1208723,\n  author       = {Daniel Serv\u00e9n and\n                  Charlie Brummitt},\n  title        = {pyGAM: Generalized Additive Models in Python},\n  month        = mar,\n  year         = 2018,\n  doi          = {10.5281/zenodo.1208723},\n  url          = {https://doi.org/10.5281/zenodo.1208723}\n}\n```\n\n## References\n1. Simon N. Wood, 2006\nGeneralized Additive Models: an introduction with R\n\n0. Hastie, Tibshirani, Friedman\nThe Elements of Statistical Learning\nhttps://www.sas.upenn.edu/~fdiebold/NoHesitations/BookAdvanced.pdf\n\n0. James, Witten, Hastie and Tibshirani\nAn Introduction to Statistical Learning\nhttp://www-bcf.usc.edu/~gareth/ISL/ISLR%20Sixth%20Printing.pdf\n\n0. Paul Eilers & Brian Marx, 1996\nFlexible Smoothing with B-splines and Penalties\nhttps://sites.stat.washington.edu/courses/stat527/s14/readings/EilersMarx_StatSci_1996.pdf\n\n0. Kim Larsen, 2015\nGAM: The Predictive Modeling Silver Bullet\nhttp://multithreaded.stitchfix.com/assets/files/gam.pdf\n\n0. Deva Ramanan, 2008\nUCI Machine Learning: Notes on IRLS\nhttp://www.ics.uci.edu/~dramanan/teaching/ics273a_winter08/homework/irls_notes.pdf\n\n0. Paul Eilers & Brian Marx, 2015\nInternational Biometric Society: A Crash Course on P-splines\nhttps://multithreaded.stitchfix.com/assets/files/gam.pdf\n\n0. Keiding, Niels, 1991\nAge-specific incidence and prevalence: a statistical perspective\nhttps://academic.oup.com/jrsssa/article-abstract/154/3/371/7106499\n\n\n<!---http://www.cs.princeton.edu/courses/archive/fall11/cos323/notes/cos323_f11_lecture09_svd.pdf--->\n\n<!---http://www.stats.uwo.ca/faculty/braun/ss3859/notes/Chapter4/ch4.pdf--->\n\n<!---http://www.stat.berkeley.edu/~census/mlesan.pdf--->\n\n<!---http://web.mit.edu/hyperbook/Patrikalakis-Maekawa-Cho/node17.html---> <!--- this helped me get spline gradients--->\n\n<!---https://scikit-sparse.readthedocs.io/en/latest/overview.html#developers--->\n\n<!---https://vincentarelbundock.github.io/Rdatasets/datasets.html---> <!--- R Datasets!--->\n",
    "bugtrack_url": null,
    "license": "Apache-2.0",
    "summary": "Generalized Additive Models in Python.",
    "version": "0.10.1",
    "project_urls": null,
    "split_keywords": [],
    "urls": [
        {
            "comment_text": null,
            "digests": {
                "blake2b_256": "bea7751c4919d7acb68abae27929a9489d52526f791ae9e3d5e32990377cd619",
                "md5": "67bd747e11cb5f4e8c209faf66064a30",
                "sha256": "9c87f644e43b9a4fa5eacddeb8a0579d35208ddf3442c23d30433b67c4f22864"
            },
            "downloads": -1,
            "filename": "pygam-0.10.1-py3-none-any.whl",
            "has_sig": false,
            "md5_digest": "67bd747e11cb5f4e8c209faf66064a30",
            "packagetype": "bdist_wheel",
            "python_version": "py3",
            "requires_python": "<3.14,>=3.9",
            "size": 80202,
            "upload_time": "2025-07-26T17:31:24",
            "upload_time_iso_8601": "2025-07-26T17:31:24.526297Z",
            "url": "https://files.pythonhosted.org/packages/be/a7/751c4919d7acb68abae27929a9489d52526f791ae9e3d5e32990377cd619/pygam-0.10.1-py3-none-any.whl",
            "yanked": false,
            "yanked_reason": null
        },
        {
            "comment_text": null,
            "digests": {
                "blake2b_256": "1162ef01ba5f1ad7225b47565e06dc52d46901812ce92fd6400a68ced17dd0b2",
                "md5": "6f901f22e53bb7761403f54635beb111",
                "sha256": "0d7c0ba235919dfadfe2347ca89c4fed5879acd428a944633959a8e0cc3e0311"
            },
            "downloads": -1,
            "filename": "pygam-0.10.1.tar.gz",
            "has_sig": false,
            "md5_digest": "6f901f22e53bb7761403f54635beb111",
            "packagetype": "sdist",
            "python_version": "source",
            "requires_python": "<3.14,>=3.9",
            "size": 78244,
            "upload_time": "2025-07-26T17:31:26",
            "upload_time_iso_8601": "2025-07-26T17:31:26.019238Z",
            "url": "https://files.pythonhosted.org/packages/11/62/ef01ba5f1ad7225b47565e06dc52d46901812ce92fd6400a68ced17dd0b2/pygam-0.10.1.tar.gz",
            "yanked": false,
            "yanked_reason": null
        }
    ],
    "upload_time": "2025-07-26 17:31:26",
    "github": false,
    "gitlab": false,
    "bitbucket": false,
    "codeberg": false,
    "lcname": "pygam"
}
        
Elapsed time: 1.02473s