Metadata-Version: 2.1
Name: qii-tool
Version: 0.1.3
Summary: A package implements Quantitative Input Influence method
Home-page: https://github.com/hovinh/QII
Author: Ho Xuan Vinh
Author-email: hxvinh.hcmus@gmail.com
License: UNKNOWN
Platform: UNKNOWN
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Description-Content-Type: text/markdown
Requires-Dist: numpy

Quantitative Input Influence - an influence measures for Algorithmic Transparency

The `qii_tool` package is an implementation of the QII method proposed in the paper ["Algorithmic Transparency via Quantitative Input Influence: Theory and Experiments with Learning Systems"](https://www.comp.nus.edu.sg/~zick/papers/transparency.oakland16.pdf). The original paper discusses the transparency-privacy tradeoff, whereas this particular package only exploits its transparency aspect to be used as an influence measures - an instance of interpretable machine learning.

Install `qii_tool` on your system using:
```python
pip install qii_tool
```
Examples can be found in this [repository](https://github.com/hovinh/QII).

