Metadata-Version: 2.4
Name: density-aware-smote
Version: 0.1.0
Summary: Density Aware SMOTE: A Python package for density-aware synthetic oversampling
Author-email: Your Name <your.email@example.com>
Project-URL: Homepage, https://github.com/nbeeeel/Improved-Oversampling-Density-Aware-Smote
Project-URL: Bug Tracker, https://github.com/nbeeeel/Improved-Oversampling-Density-Aware-Smote/issues
Keywords: SMOTE,oversampling,machine-learning,imbalanced-data,synthetic-data
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Science/Research
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.8
Classifier: Programming Language :: Python :: 3.9
Classifier: Programming Language :: Python :: 3.10
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Python: >=3.7
Description-Content-Type: text/markdown
License-File: LICENSE.txt
Requires-Dist: numpy
Requires-Dist: scikit-learn
Requires-Dist: matplotlib
Dynamic: license-file

# Density Aware SMOTE

![PyPI version](https://img.shields.io/pypi/v/density-aware-smote?color=blue)
![Python versions](https://img.shields.io/pypi/pyversions/density-aware-smote)
![License](https://img.shields.io/github/license/nbeeeel/Improved-Oversampling-Density-Aware-Smote)

A Python package implementing **Density Aware SMOTE**, developed in the [**Improved-Oversampling-Density-Aware-Smote** repository](https://github.com/nbeeeel/Improved-Oversampling-Density-Aware-Smote).  
This method enhances SMOTE by generating synthetic samples based on **local data density**, reducing oversampling in sparse regions and preserving valuable patterns in dense areas.

---

##  Features
- Density-aware oversampling tuned for local distribution  
- Flexible sampling strategy: `auto`, ratio float, or custom class counts  
- Custom neighbor selection: `random`, `nearest`, `farthest`  
- Seamless integration with scikit-learn pipelines  
- Built-in visuals: class distribution, synthetic samples, decision boundaries  

---

##  Installation

```bash
pip install density-aware-smote
