Metadata-Version: 2.4
Name: safe-ai-metrics
Version: 0.1.1
Summary: SAFE-AI metrics for accuracy, robustness, and explainability evaluation
Author: Vasily Kolesnikov
License-Expression: MIT
Keywords: safe-ai,robustness,explainability,machine-learning,metrics,model-risk,responsible-ai,trustworthy-ai
Classifier: Development Status :: 3 - Alpha
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Intended Audience :: Developers
Classifier: Intended Audience :: Science/Research
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Classifier: Topic :: Scientific/Engineering :: Information Analysis
Requires-Python: >=3.10
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: numpy>=1.24
Requires-Dist: matplotlib>=3.7
Requires-Dist: scikit-learn>=1.3
Requires-Dist: torch>=2.0
Requires-Dist: torchvision>=0.15
Requires-Dist: opencv-python>=4.8
Requires-Dist: Pillow>=10
Requires-Dist: adversarial-robustness-toolbox>=1.18
Dynamic: license-file

# SAFE-AI Metrics

SAFE-AI Metrics is a Python package for evaluating machine learning models using accuracy, robustness, and explainability-oriented metrics.

The package provides implementations of:

- **RGA** — Rank Graduation Accuracy
- **RGE** — Rank Graduation Explainability
- **RGR** — Rank Graduation Robustness

## Installation

```bash
pip install safe-ai-metrics
```

## Acknowledgements

The development of this package builds on the `safeaipackage` project by Golnoosh Babaei.

The original `safeaipackage` repository is available at: https://github.com/GolnooshBabaei/safeaipackage

This repository is currently maintained as a separate implementation for development and packaging purposes, but it is expected to be merged or aligned with the original SAFE-AI package in the future.

If you use this package in academic work, please consider citing the related SAFE-AI and RGB paper referenced in the original project.
