# FeatBoost-X
Python implementation of FeatBoost-X. See the [paper]() for details.
## Usage
```shell
pip install featboostx
```
### Example
```python
from featboostx import FeatBoostClassifier
clf = FeatBoostClassifier()
clf.fit(X, y)
print(clf.selected_subset_)
```
For a more detailed example, see the [classification example](examples/example_classification.py) or the
[regression example](examples/example_regression.py).
## Feature selection methods
FeatBoost-X is available classification, regression, and survival problems.
- Classification supports the objectives accuracy (`acc`) and the F1-score (`f1`) through the `FeatBoostClassifier`-class. These can be optimized through the `softmax` or `adaboost` objective.
This implementation originates from the Python implementation of the [original paper](https://github.com/amjams/FeatBoost).
- Regression supports the `mae` objective through the `FeatBoostRegressor`-class and can be optimized through `adaptive` boosting.
- Survival supports the `c_index` objective through the `FeatBoostRegressor`-class and can be optimized through `adaptive` boosting.
# Illustration of FeatBoost-X
Raw data
{
"_id": null,
"home_page": "https://github.com/O-T-O-Z/FeatBoost-X",
"name": "featboostx",
"maintainer": null,
"docs_url": null,
"requires_python": "<4,>=3.10",
"maintainer_email": null,
"keywords": "Feature Selection, Gradient Boosting, FeatBoost-X",
"author": "\u00d6mer Tarik \u00d6zyilmaz & Ahmad Alsahaf",
"author_email": "o.t.ozyilmaz@umcg.nl",
"download_url": "https://files.pythonhosted.org/packages/c8/fc/a4bd50e2563d15e6938a1eb8ce4dfad00b41e8127eda16f8ad949ec58040/featboostx-0.1.0rc2.tar.gz",
"platform": null,
"description": "# FeatBoost-X\nPython implementation of FeatBoost-X. See the [paper]() for details.\n\n\n## Usage\n```shell\npip install featboostx\n```\n\n### Example\n```python\nfrom featboostx import FeatBoostClassifier\n\nclf = FeatBoostClassifier()\nclf.fit(X, y)\nprint(clf.selected_subset_)\n```\nFor a more detailed example, see the [classification example](examples/example_classification.py) or the \n[regression example](examples/example_regression.py).\n\n## Feature selection methods\nFeatBoost-X is available classification, regression, and survival problems.\n- Classification supports the objectives accuracy (`acc`) and the F1-score (`f1`) through the `FeatBoostClassifier`-class. These can be optimized through the `softmax` or `adaboost` objective.\nThis implementation originates from the Python implementation of the [original paper](https://github.com/amjams/FeatBoost).\n- Regression supports the `mae` objective through the `FeatBoostRegressor`-class and can be optimized through `adaptive` boosting.\n- Survival supports the `c_index` objective through the `FeatBoostRegressor`-class and can be optimized through `adaptive` boosting.\n\n\n# Illustration of FeatBoost-X\n\n",
"bugtrack_url": null,
"license": null,
"summary": "A Python package for the FeatBoost-X feature selection algorithm",
"version": "0.1.0rc2",
"project_urls": {
"Homepage": "https://github.com/O-T-O-Z/FeatBoost-X"
},
"split_keywords": [
"feature selection",
" gradient boosting",
" featboost-x"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "c8a6fb1441db33f17edd762a33d0c55e6c83952394853ebd7575395908e84b6a",
"md5": "36b69b6fde68f5f49925cf318d45b835",
"sha256": "f037834245e50f6a5dd7be3c12dd75e1078b4940bca63cd8b6a11d05fedfc05f"
},
"downloads": -1,
"filename": "featboostx-0.1.0rc2-py3-none-any.whl",
"has_sig": false,
"md5_digest": "36b69b6fde68f5f49925cf318d45b835",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": "<4,>=3.10",
"size": 12842,
"upload_time": "2025-01-10T13:26:30",
"upload_time_iso_8601": "2025-01-10T13:26:30.607459Z",
"url": "https://files.pythonhosted.org/packages/c8/a6/fb1441db33f17edd762a33d0c55e6c83952394853ebd7575395908e84b6a/featboostx-0.1.0rc2-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "c8fca4bd50e2563d15e6938a1eb8ce4dfad00b41e8127eda16f8ad949ec58040",
"md5": "38c0b4521872441d84a1949425577bf1",
"sha256": "95e24390630baed299ead5b0c7bb2f6e4ad3de7d5b5b2c67b5a5f5bcabe9835e"
},
"downloads": -1,
"filename": "featboostx-0.1.0rc2.tar.gz",
"has_sig": false,
"md5_digest": "38c0b4521872441d84a1949425577bf1",
"packagetype": "sdist",
"python_version": "source",
"requires_python": "<4,>=3.10",
"size": 366150,
"upload_time": "2025-01-10T13:26:37",
"upload_time_iso_8601": "2025-01-10T13:26:37.169839Z",
"url": "https://files.pythonhosted.org/packages/c8/fc/a4bd50e2563d15e6938a1eb8ce4dfad00b41e8127eda16f8ad949ec58040/featboostx-0.1.0rc2.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-01-10 13:26:37",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "O-T-O-Z",
"github_project": "FeatBoost-X",
"travis_ci": false,
"coveralls": false,
"github_actions": false,
"requirements": [
{
"name": "numpy",
"specs": [
[
"==",
"1.26.4"
]
]
},
{
"name": "shap",
"specs": [
[
"==",
"0.45.0"
]
]
},
{
"name": "lifelines",
"specs": [
[
"==",
"0.28.0"
]
]
},
{
"name": "pytest",
"specs": [
[
"==",
"8.3.3"
]
]
},
{
"name": "xgboost",
"specs": [
[
"==",
"2.0.3"
]
]
},
{
"name": "scipy",
"specs": [
[
"==",
"1.13"
]
]
}
],
"lcname": "featboostx"
}
JSON
