# Regressionizer
Python package with a class that allows pipeline-like specification and execution of regression workflows.
Extensive guide is given in the Jupyter notebook
["Rapid-specification-of-regression-workflows.ipynb"](https://github.com/antononcube/Python-Regressionizer/blob/main/docs/Rapid-specification-of-regression-workflows.ipynb)
and the corresponding Markdown document
["Rapid-specification-of-regression-workflows.md"](https://github.com/antononcube/Python-Regressionizer/blob/main/docs/Rapid-specification-of-regression-workflows.md).
------
## Features summary
- The class `Regressionizer` facilitates rapid specifications of regressions workflows.
- To quickly specify:
- data rescaling and summary
- regression computations
- outliers finding
- conditional Cumulative Distribution Functions (CDFs) reconstruction
- plotting of data, fits, residual errors, outliers, CDFs
- `Regressionizer` works with data frames, numpy arrays, lists of numbers, and lists of numeric pairs.
### Details and arguments
- The curves computed with Quantile Regression are called **regression quantiles**.
- `Regressionizer` has three regression methods:
- `quantile_regression`
- `quantile_regression_fit`
- `least_squares_fit`
- The regression quantiles computed with the methods `quantile_regression` and `quantile_regression_fit`
correspond to probabilities specified with the argument `probs`.
- The method`quantile_regression` computes fits using a B-spline functions basis.
- The basis is specified with the arguments `knots` and `order`.
- `order` is 3 by default.
- The methods `quantile_regession_fit` and `least_squares_fit` use lists of basis functions to fit with
specified with the argument `funcs`.
### Workflows flowchart
The following flowchart summarizes the workflows that are supported by `Regressionizer`:
------
## Usage examples
Import libraries:
```python
from Regressionizer import *
import numpy as np
```
Generate random data:
```python
np.random.seed(0)
x = np.linspace(0, 2, 300)
y = np.sin(2 * np.pi * x) + np.random.normal(0, 0.4, x.shape)
data = np.column_stack((x, y)
```
Compute quantile regression for probabilities `[0.2, 0.5, 0.8]` and make the corresponding plot:
```python
obj = (Regressionizer(data)
.quantile_regression(knots=8, probs=[0.2, 0.5, 0.8])
.plot(title="B-splines fit", template="plotly")
)
```
Show the plot obtained above:
```python
obj.take_value().show()
```
------
## References
### Articles, books
[RK1] Roger Koenker,
[Quantile Regression](https://books.google.com/books/about/Quantile_Regression.html?id=hdkt7V4NXsgC),
Cambridge University Press, 2005.
[RK2] Roger Koenker,
["Quantile Regression in R: a vignette"](https://cran.r-project.org/web/packages/quantreg/vignettes/rq.pdf),
(2006),
[CRAN](https://cran.r-project.org/).
[AA1] Anton Antonov,
["A monad for Quantile Regression workflows"](https://github.com/antononcube/MathematicaForPrediction/blob/master/MarkdownDocuments/A-monad-for-Quantile-Regression-workflows.md),
(2018),
[MathematicaForPrediction at GitHub](https://github.com/antononcube/MathematicaForPrediction).
### Packages, paclets
[RKp1] Roger Koenker,
[`quantreg`](https://cran.r-project.org/web/packages/quantreg/index.html),
[CRAN](https://cran.r-project.org/).
[AAp1] Anton Antonov,
[Quantile Regression WL paclet](https://github.com/antononcube/WL-QuantileRegression-paclet),
(2014-2023),
[GitHub/antononcube](https://github.com/antononcube).
[AAp2] Anton Antonov,
[Monadic Quantile Regression WL paclet](https://github.com/antononcube/WL-MonadicQuantileRegression-paclet),
(2018-2024),
[GitHub/antononcube](https://github.com/antononcube).
[AAp3] Anton Antonov,
[`QuantileRegression`](https://resources.wolframcloud.com/FunctionRepository/resources/QuantileRegression),
(2019),
[Wolfram Function Repository](https://resources.wolframcloud.com/FunctionRepository/resources/QuantileRegression).
### Repositories
[AAr1] Anton Antonov,
[DSL::English::QuantileRegressionWorkflows in Raku](https://github.com/antononcube/Raku-DSL-English-QuantileRegressionWorkflows),
(2020),
[GitHub/antononcube](https://github.com/antononcube/Raku-DSL-English-QuantileRegressionWorkflows).
### Videos
[AAv1] Anton Antonov,
["Boston useR! QuantileRegression Workflows 2019-04-18"](https://www.youtube.com/watch?v=a_Dk25xarvE),
(2019),
[Anton Antonov at YouTube](https://www.youtube.com/@AAA4Prediction).
[AAv2] Anton Antonov,
["useR! 2020: How to simplify Machine Learning workflows specifications"](https://www.youtube.com/watch?v=b9Uu7gRF5KY),
(2020),
[R Consortium at YouTube](https://www.youtube.com/channel/UC_R5smHVXRYGhZYDJsnXTwg).
Raw data
{
"_id": null,
"home_page": "https://github.com/antononcube/Python-Regressionizer",
"name": "Regressionizer",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.7",
"maintainer_email": null,
"keywords": "regression, quantile, quantile regression, linear, linear regression, workflow",
"author": "Anton Antonov",
"author_email": "antononcube@posteo.net",
"download_url": "https://files.pythonhosted.org/packages/ea/3d/8062cee8b68346a964224e8338283fca81930cb9a7ea3121a47a65490b7c/regressionizer-0.1.9.tar.gz",
"platform": null,
"description": "# Regressionizer\n\nPython package with a class that allows pipeline-like specification and execution of regression workflows.\n\nExtensive guide is given in the Jupyter notebook \n[\"Rapid-specification-of-regression-workflows.ipynb\"](https://github.com/antononcube/Python-Regressionizer/blob/main/docs/Rapid-specification-of-regression-workflows.ipynb) \nand the corresponding Markdown document\n[\"Rapid-specification-of-regression-workflows.md\"](https://github.com/antononcube/Python-Regressionizer/blob/main/docs/Rapid-specification-of-regression-workflows.md).\n\n------\n\n## Features summary \n\n- The class `Regressionizer` facilitates rapid specifications of regressions workflows.\n - To quickly specify: \n - data rescaling and summary\n - regression computations\n - outliers finding\n - conditional Cumulative Distribution Functions (CDFs) reconstruction\n - plotting of data, fits, residual errors, outliers, CDFs \n\n- `Regressionizer` works with data frames, numpy arrays, lists of numbers, and lists of numeric pairs.\n\n\n### Details and arguments\n\n- The curves computed with Quantile Regression are called **regression quantiles**.\n\n- `Regressionizer` has three regression methods:\n - `quantile_regression`\n - `quantile_regression_fit`\n - `least_squares_fit`\n \n- The regression quantiles computed with the methods `quantile_regression` and `quantile_regression_fit` \n correspond to probabilities specified with the argument `probs`.\n\n- The method`quantile_regression` computes fits using a B-spline functions basis. \n - The basis is specified with the arguments `knots` and `order`.\n - `order` is 3 by default. \n\n- The methods `quantile_regession_fit` and `least_squares_fit` use lists of basis functions to fit with \n specified with the argument `funcs`.\n\n### Workflows flowchart\n\nThe following flowchart summarizes the workflows that are supported by `Regressionizer`:\n\n\n\n\n------\n\n## Usage examples \n\nImport libraries:\n\n```python\nfrom Regressionizer import *\nimport numpy as np\n```\n\nGenerate random data:\n\n```python\nnp.random.seed(0)\nx = np.linspace(0, 2, 300)\ny = np.sin(2 * np.pi * x) + np.random.normal(0, 0.4, x.shape)\ndata = np.column_stack((x, y)\n```\n\nCompute quantile regression for probabilities `[0.2, 0.5, 0.8]` and make the corresponding plot:\n\n```python\nobj = (Regressionizer(data)\n .quantile_regression(knots=8, probs=[0.2, 0.5, 0.8])\n .plot(title=\"B-splines fit\", template=\"plotly\")\n )\n```\n\nShow the plot obtained above:\n\n```python\nobj.take_value().show()\n```\n\n\n\n------\n\n## References\n\n### Articles, books\n\n[RK1] Roger Koenker, \n[Quantile Regression](https://books.google.com/books/about/Quantile_Regression.html?id=hdkt7V4NXsgC), \nCambridge University Press, 2005.\n\n[RK2] Roger Koenker,\n[\"Quantile Regression in R: a vignette\"](https://cran.r-project.org/web/packages/quantreg/vignettes/rq.pdf),\n(2006),\n[CRAN](https://cran.r-project.org/).\n\n[AA1] Anton Antonov,\n[\"A monad for Quantile Regression workflows\"](https://github.com/antononcube/MathematicaForPrediction/blob/master/MarkdownDocuments/A-monad-for-Quantile-Regression-workflows.md),\n(2018),\n[MathematicaForPrediction at GitHub](https://github.com/antononcube/MathematicaForPrediction).\n\n### Packages, paclets\n\n[RKp1] Roger Koenker,\n[`quantreg`](https://cran.r-project.org/web/packages/quantreg/index.html),\n[CRAN](https://cran.r-project.org/).\n\n[AAp1] Anton Antonov,\n[Quantile Regression WL paclet](https://github.com/antononcube/WL-QuantileRegression-paclet),\n(2014-2023),\n[GitHub/antononcube](https://github.com/antononcube).\n\n[AAp2] Anton Antonov,\n[Monadic Quantile Regression WL paclet](https://github.com/antononcube/WL-MonadicQuantileRegression-paclet),\n(2018-2024),\n[GitHub/antononcube](https://github.com/antononcube).\n\n[AAp3] Anton Antonov,\n[`QuantileRegression`](https://resources.wolframcloud.com/FunctionRepository/resources/QuantileRegression),\n(2019),\n[Wolfram Function Repository](https://resources.wolframcloud.com/FunctionRepository/resources/QuantileRegression).\n\n### Repositories\n\n[AAr1] Anton Antonov,\n[DSL::English::QuantileRegressionWorkflows in Raku](https://github.com/antononcube/Raku-DSL-English-QuantileRegressionWorkflows),\n(2020),\n[GitHub/antononcube](https://github.com/antononcube/Raku-DSL-English-QuantileRegressionWorkflows).\n\n### Videos\n\n[AAv1] Anton Antonov,\n[\"Boston useR! QuantileRegression Workflows 2019-04-18\"](https://www.youtube.com/watch?v=a_Dk25xarvE),\n(2019),\n[Anton Antonov at YouTube](https://www.youtube.com/@AAA4Prediction).\n\n[AAv2] Anton Antonov,\n[\"useR! 2020: How to simplify Machine Learning workflows specifications\"](https://www.youtube.com/watch?v=b9Uu7gRF5KY),\n(2020),\n[R Consortium at YouTube](https://www.youtube.com/channel/UC_R5smHVXRYGhZYDJsnXTwg).\n",
"bugtrack_url": null,
"license": null,
"summary": "Regression workflows package based on Least Squares Regression and Quantile Regression.",
"version": "0.1.9",
"project_urls": {
"Homepage": "https://github.com/antononcube/Python-Regressionizer"
},
"split_keywords": [
"regression",
" quantile",
" quantile regression",
" linear",
" linear regression",
" workflow"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "852903dcdf644298616c61ace4fb28961125c2ab422d6e4c285876b93142744b",
"md5": "1ec9132644c05bf3def918541e6b7e02",
"sha256": "ed487784ce58b783e21ff91bb92c74555c20a4bc14a577e70fdf2352dd876737"
},
"downloads": -1,
"filename": "Regressionizer-0.1.9-py3-none-any.whl",
"has_sig": false,
"md5_digest": "1ec9132644c05bf3def918541e6b7e02",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.7",
"size": 22791,
"upload_time": "2024-12-07T16:11:18",
"upload_time_iso_8601": "2024-12-07T16:11:18.436309Z",
"url": "https://files.pythonhosted.org/packages/85/29/03dcdf644298616c61ace4fb28961125c2ab422d6e4c285876b93142744b/Regressionizer-0.1.9-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "ea3d8062cee8b68346a964224e8338283fca81930cb9a7ea3121a47a65490b7c",
"md5": "6a29437abdfac9a71064102aca458502",
"sha256": "75009e04708f22bbd738ac1e3b32ad1bdc900a8fac1a6017b9f7a30dc38dd41d"
},
"downloads": -1,
"filename": "regressionizer-0.1.9.tar.gz",
"has_sig": false,
"md5_digest": "6a29437abdfac9a71064102aca458502",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.7",
"size": 24571,
"upload_time": "2024-12-07T16:11:19",
"upload_time_iso_8601": "2024-12-07T16:11:19.594068Z",
"url": "https://files.pythonhosted.org/packages/ea/3d/8062cee8b68346a964224e8338283fca81930cb9a7ea3121a47a65490b7c/regressionizer-0.1.9.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-12-07 16:11:19",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "antononcube",
"github_project": "Python-Regressionizer",
"travis_ci": false,
"coveralls": false,
"github_actions": false,
"lcname": "regressionizer"
}
JSON
