<!-- HEADER -->
| | |
|---|---|
| <img src="https://www.ieseg.fr/wp-content/uploads/IESEG-Logo-2012-rgb.jpg" alt="drawing" width=100%/> | <span><br>Credit Scoring<br>Module<br>Class: 2022 & 2023</span> |
<!-- CONTENT -->
---
## Overview
- Odds based Grouping (OBG)
- OBGEncoder
- `pred_var`: Name of predictor variable. Values can be either continuous or categorical.
- `target_var`: Name of binary target variable.
.fit
- `df`: DataFrame containing pred_var and target_var.
- `max_delta`: max difference between odds for merging two levels. default: 0.05
- `min_bins`: minimum number of bins. default: 3
- `q`: number of quantiles when converting continuous variable to categorical. default: 10
.transform
- `df`: Transform pred_var based on fitted bins.
- `impute`: Boolean indicating whether to impute missing values. default: False
- `impute_value`: Category level to impute missing values with. default: 'Missing' or 'nan'
.fit_transform
- `df`: DataFrame containing pred_var and target_var. Transform pred_var based on fitted bins.
>fit_dict: dictionary containing the matched category levels and fitted bins.
>lookup: dictionary containing cutoff values for continuous variable (empty if pred_var is categorical).
- Weight of Evidence (WOE)
- WOEEncoder
- `pred_var`: Name of predictor variable.Values can be either continuous or categorical.
- `target_var`: Name of binary target variable.
- `target_value`: Value indicating event. default: 1.
.fit
- `df`: DataFrame containing pred_var and target_var.
- `stop_limit`: Stops WOE based merging of the predictor's classes/levels in case the resulting information value (IV) decreases more than (e.g. 0.05 = 5%) compared to the preceding binning step. stop_limit=0 will skip any WOE based merging. Increasing the stop_limit will simplify the binning solution and may avoid overfitting. Accepted value range: 0 to 0.5. default: 0.1.
- `q`: number of quantiles when converting continuous variable to categorical. default: 10
.transform
- `df`: Transform pred_var based on fitted bins
- `impute`: Boolean indicating whether to impute missing values. default: False
- `impute_value`: Category level to impute missing values with. default: 'Missing' or 'nan'
.fit_transform
- `df`: DataFrame containing pred_var and target_var. Transform pred_var based on fitted bins.
.test_limit
- `df`: DataFrame containing pred_var and target_var to test stop limits at 1%, 2.5%, 5% and 10%.
>fit_dict: dictionary containing the matched category levels and fitted bins.
>lookup: dictionary containing cutoff values for continuous variable (empty if pred_var is categorical).
<br>
Raw data
{
"_id": null,
"home_page": "https://github.com/pnborchert",
"name": "ieseg-credscore",
"maintainer": "",
"docs_url": null,
"requires_python": "",
"maintainer_email": "",
"keywords": "Credit Scoring IESEG",
"author": "Philipp Borchert",
"author_email": "p.borchert@ieseg.fr",
"download_url": "https://files.pythonhosted.org/packages/f4/5f/1aea2549175421123f37a37777ed2d28c5592ae1ce04150f3efeb4bda9e4/ieseg_credscore-0.23.12.tar.gz",
"platform": null,
"description": "<!-- HEADER -->\r\n| | |\r\n|---|---|\r\n| <img src=\"https://www.ieseg.fr/wp-content/uploads/IESEG-Logo-2012-rgb.jpg\" alt=\"drawing\" width=100%/> | <span><br>Credit Scoring<br>Module<br>Class: 2022 & 2023</span> |\r\n\r\n<!-- CONTENT -->\r\n\r\n---\r\n\r\n## Overview\r\n\r\n- Odds based Grouping (OBG)\r\n - OBGEncoder\r\n\r\n - `pred_var`: Name of predictor variable. Values can be either continuous or categorical.\r\n - `target_var`: Name of binary target variable.\r\n\r\n .fit\r\n\r\n - `df`: DataFrame containing pred_var and target_var.\r\n - `max_delta`: max difference between odds for merging two levels. default: 0.05\r\n - `min_bins`: minimum number of bins. default: 3\r\n - `q`: number of quantiles when converting continuous variable to categorical. default: 10\r\n\r\n .transform\r\n\r\n - `df`: Transform pred_var based on fitted bins.\r\n - `impute`: Boolean indicating whether to impute missing values. default: False\r\n - `impute_value`: Category level to impute missing values with. default: 'Missing' or 'nan'\r\n\r\n .fit_transform\r\n\r\n - `df`: DataFrame containing pred_var and target_var. Transform pred_var based on fitted bins.\r\n\r\n >fit_dict: dictionary containing the matched category levels and fitted bins.\r\n\r\n >lookup: dictionary containing cutoff values for continuous variable (empty if pred_var is categorical).\r\n\r\n\r\n\r\n- Weight of Evidence (WOE)\r\n\r\n - WOEEncoder\r\n\r\n - `pred_var`: Name of predictor variable.Values can be either continuous or categorical.\r\n - `target_var`: Name of binary target variable. \r\n - `target_value`: Value indicating event. default: 1.\r\n\r\n .fit\r\n\r\n - `df`: DataFrame containing pred_var and target_var.\r\n - `stop_limit`: Stops WOE based merging of the predictor's classes/levels in case the resulting information value (IV) decreases more than (e.g. 0.05 = 5%) compared to the preceding binning step. stop_limit=0 will skip any WOE based merging. Increasing the stop_limit will simplify the binning solution and may avoid overfitting. Accepted value range: 0 to 0.5. default: 0.1.\r\n - `q`: number of quantiles when converting continuous variable to categorical. default: 10\r\n\r\n .transform\r\n\r\n - `df`: Transform pred_var based on fitted bins\r\n - `impute`: Boolean indicating whether to impute missing values. default: False\r\n - `impute_value`: Category level to impute missing values with. default: 'Missing' or 'nan'\r\n\r\n .fit_transform\r\n\r\n - `df`: DataFrame containing pred_var and target_var. Transform pred_var based on fitted bins.\r\n\r\n .test_limit\r\n\r\n - `df`: DataFrame containing pred_var and target_var to test stop limits at 1%, 2.5%, 5% and 10%.\r\n\r\n >fit_dict: dictionary containing the matched category levels and fitted bins.\r\n\r\n >lookup: dictionary containing cutoff values for continuous variable (empty if pred_var is categorical).\r\n\r\n<br>\r\n",
"bugtrack_url": null,
"license": "MIT",
"summary": "Credit Scoring - IESEG School of Management",
"version": "0.23.12",
"project_urls": {
"Homepage": "https://github.com/pnborchert"
},
"split_keywords": [
"credit",
"scoring",
"ieseg"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "f45f1aea2549175421123f37a37777ed2d28c5592ae1ce04150f3efeb4bda9e4",
"md5": "3236129cf36818fe20c7010710589faa",
"sha256": "e5267a48aa9837634558d977c6bd130745e2dcd0c8319eae78a4f478ac673a1d"
},
"downloads": -1,
"filename": "ieseg_credscore-0.23.12.tar.gz",
"has_sig": false,
"md5_digest": "3236129cf36818fe20c7010710589faa",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 12198,
"upload_time": "2023-07-05T09:27:51",
"upload_time_iso_8601": "2023-07-05T09:27:51.545145Z",
"url": "https://files.pythonhosted.org/packages/f4/5f/1aea2549175421123f37a37777ed2d28c5592ae1ce04150f3efeb4bda9e4/ieseg_credscore-0.23.12.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2023-07-05 09:27:51",
"github": false,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"lcname": "ieseg-credscore"
}
JSON