| Name | splinator JSON |
| Version |
0.2.0
JSON |
| download |
| home_page | |
| Summary | Python library for fitting linear-spine based logistic regression for calibration. |
| upload_time | 2023-07-31 07:47:27 |
| maintainer | |
| docs_url | None |
| author | |
| requires_python | <4.0,>=3.7.1 |
| license | BSD-3-Clause |
| keywords |
calibration
logistic
spline
regression
|
| VCS |
|
| bugtrack_url |
|
| requirements |
No requirements were recorded.
|
| Travis-CI |
No Travis.
|
| coveralls test coverage |
No coveralls.
|
# Splinator 📈
**Probablistic Calibration with Regression Splines**
[scikit-learn](https://scikit-learn.org) compatible
[](https://pdm.fming.dev)
[](https://splinator.readthedocs.io/en/latest/)
[](https://github.com/affirm/splinator/actions)
## Installation
`pip install splinator`
## Algorithm
Supported models:
- Linear Spline Logistic Regression
Supported metrics:
- Spiegelhalter’s z statistic
- Expected Calibration Error (ECE)
\[1\] You can find more information in the [Linear Spline Logistic
Regression](https://github.com/Affirm/splinator/wiki/Linear-Spline-Logistic-Regression).
\[2\] Additional readings
- Zhang, Jian, and Yiming Yang. [Probabilistic score estimation with
piecewise logistic
regression](https://pal.sri.com/wp-content/uploads/publications/radar/2004/icml04zhang.pdf).
Proceedings of the twenty-first international conference on Machine
learning. 2004.
- Guo, Chuan, et al. "On calibration of modern neural networks." International conference on machine learning. PMLR, 2017.
## Examples
| comparison | notebook |
|------------------------------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| scikit-learn's sigmoid and isotonic regression | [](https://github.com/Affirm/splinator/blob/main/examples/calibrator_model_comparison.ipynb) |
| pyGAM’s spline model | [](https://githubtocolab.com/Affirm/splinator/blob/main/examples/spline_model_comparison.ipynb) |
## Development
The dependencies are managed by [pdm](https://pdm.fming.dev/latest/)
To run tests, run `pdm run -v pytest tests`
## Example Usage
``` python
from splinator.estimators import LinearSplineLogisticRegression
import numpy as np
# random synthetic dataset
n_samples = 100
rng = np.random.RandomState(0)
X = rng.normal(loc=100, size=(n_samples, 2))
y = np.random.randint(2, size=n_samples)
lslr = LinearSplineLogisticRegression(n_knots=10)
lslr.fit(X, y)
```
Raw data
{
"_id": null,
"home_page": "",
"name": "splinator",
"maintainer": "",
"docs_url": null,
"requires_python": "<4.0,>=3.7.1",
"maintainer_email": "",
"keywords": "calibration logistic spline regression",
"author": "",
"author_email": "Jiarui Xu <jiarui.xu@affirm.com>",
"download_url": "https://files.pythonhosted.org/packages/65/18/93b83d9923e1b3c352d589faf3cafad5641f7ea6f4d5400be37efe70ecbd/splinator-0.2.0.tar.gz",
"platform": null,
"description": "# Splinator \ud83d\udcc8\n\n**Probablistic Calibration with Regression Splines**\n\n[scikit-learn](https://scikit-learn.org) compatible\n\n[](https://pdm.fming.dev)\n[](https://splinator.readthedocs.io/en/latest/)\n[](https://github.com/affirm/splinator/actions)\n\n## Installation\n\n`pip install splinator`\n\n## Algorithm\n\nSupported models:\n\n- Linear Spline Logistic Regression\n\nSupported metrics:\n\n- Spiegelhalter\u2019s z statistic\n- Expected Calibration Error (ECE)\n\n\\[1\\] You can find more information in the [Linear Spline Logistic\nRegression](https://github.com/Affirm/splinator/wiki/Linear-Spline-Logistic-Regression).\n\n\\[2\\] Additional readings\n\n- Zhang, Jian, and Yiming Yang. [Probabilistic score estimation with\n piecewise logistic\n regression](https://pal.sri.com/wp-content/uploads/publications/radar/2004/icml04zhang.pdf).\n Proceedings of the twenty-first international conference on Machine\n learning. 2004.\n- Guo, Chuan, et al. \"On calibration of modern neural networks.\" International conference on machine learning. PMLR, 2017.\n\n\n## Examples\n\n| comparison | notebook |\n|------------------------------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------|\n| scikit-learn's sigmoid and isotonic regression | [](https://github.com/Affirm/splinator/blob/main/examples/calibrator_model_comparison.ipynb) |\n| pyGAM\u2019s spline model | [](https://githubtocolab.com/Affirm/splinator/blob/main/examples/spline_model_comparison.ipynb) |\n\n## Development\n\nThe dependencies are managed by [pdm](https://pdm.fming.dev/latest/)\n\nTo run tests, run `pdm run -v pytest tests`\n\n## Example Usage\n\n``` python\nfrom splinator.estimators import LinearSplineLogisticRegression\nimport numpy as np\n\n# random synthetic dataset\nn_samples = 100\nrng = np.random.RandomState(0)\nX = rng.normal(loc=100, size=(n_samples, 2))\ny = np.random.randint(2, size=n_samples)\n\nlslr = LinearSplineLogisticRegression(n_knots=10)\nlslr.fit(X, y)\n```\n",
"bugtrack_url": null,
"license": "BSD-3-Clause",
"summary": "Python library for fitting linear-spine based logistic regression for calibration.",
"version": "0.2.0",
"project_urls": null,
"split_keywords": [
"calibration",
"logistic",
"spline",
"regression"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "21751d5d9bf6d7873ac25df91bd073c64bfb30928510c785e43cd629441c6c02",
"md5": "0dea71c33e6cb8e5e2fe0d3c50aac9ef",
"sha256": "52670b0ab8f40847d3f1abec625005e7e121efd65fae4b3af5fe1919ee36e8d2"
},
"downloads": -1,
"filename": "splinator-0.2.0-py3-none-any.whl",
"has_sig": false,
"md5_digest": "0dea71c33e6cb8e5e2fe0d3c50aac9ef",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": "<4.0,>=3.7.1",
"size": 9988,
"upload_time": "2023-07-31T07:47:25",
"upload_time_iso_8601": "2023-07-31T07:47:25.977175Z",
"url": "https://files.pythonhosted.org/packages/21/75/1d5d9bf6d7873ac25df91bd073c64bfb30928510c785e43cd629441c6c02/splinator-0.2.0-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "651893b83d9923e1b3c352d589faf3cafad5641f7ea6f4d5400be37efe70ecbd",
"md5": "aa114a003d3bcaf44a196e5fe6253d8c",
"sha256": "dc0fabd7e4a57eb7bffebf3c16901e8f5e400aa1b4c1a3b40b2830a4d4cb7caf"
},
"downloads": -1,
"filename": "splinator-0.2.0.tar.gz",
"has_sig": false,
"md5_digest": "aa114a003d3bcaf44a196e5fe6253d8c",
"packagetype": "sdist",
"python_version": "source",
"requires_python": "<4.0,>=3.7.1",
"size": 13929,
"upload_time": "2023-07-31T07:47:27",
"upload_time_iso_8601": "2023-07-31T07:47:27.628152Z",
"url": "https://files.pythonhosted.org/packages/65/18/93b83d9923e1b3c352d589faf3cafad5641f7ea6f4d5400be37efe70ecbd/splinator-0.2.0.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2023-07-31 07:47:27",
"github": false,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"lcname": "splinator"
}
