Metadata-Version: 2.4
Name: predykt
Version: 0.1.1
Summary: A rigorous Python toolkit for predictive ML — feature analysis, interaction testing, and model robustness validation
Home-page: https://github.com/HishamSalem/predykt
Author: Hisham Salem
Author-email: hisham.salem@mail.mcgill.ca
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Requires-Python: >=3.8
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: numpy
Requires-Dist: pandas
Requires-Dist: scikit-learn
Requires-Dist: shap
Requires-Dist: optbinning
Requires-Dist: scipy
Requires-Dist: statsmodels
Requires-Dist: tqdm
Requires-Dist: matplotlib
Requires-Dist: seaborn
Requires-Dist: joblib
Requires-Dist: numba
Provides-Extra: salib
Requires-Dist: SALib; extra == "salib"
Provides-Extra: xgboost
Requires-Dist: xgboost; extra == "xgboost"
Provides-Extra: full
Requires-Dist: SALib; extra == "full"
Requires-Dist: xgboost; extra == "full"
Dynamic: author
Dynamic: author-email
Dynamic: classifier
Dynamic: description
Dynamic: description-content-type
Dynamic: home-page
Dynamic: license-file
Dynamic: provides-extra
Dynamic: requires-dist
Dynamic: requires-python
Dynamic: summary

# predykt

A comprehensive Python toolkit for analyzing feature interactions in machine learning models, combining multiple methodologies to provide deep insights into feature relationships and their impact on model behavior.

## Features

### Interaction Analysis Methods
- **SHAP Interaction Analysis**: Leverages SHAP values to detect and quantify feature interactions
- **Feature Binning Analysis**: Uses optimal binning techniques to identify non-linear relationships

### Key Capabilities
- Statistical significance testing for interactions
- Visualization of interaction effects
- Multiple testing correction

## Installation

```bash
pip install predykt
