Jax-based Recourse Explanation Library
Project description
ReLax
Overview | Installation | Tutorials | Documentation | Citing ReLax
Overview
ReLax (Recourse Explanation Library in Jax) is a library
built on top of jax to generate counterfactual and recourse
explanations. By leveraging vectorization though vmap/pmap and
just-in-time compilation in jax, ReLax offers massive speed
improvements in generating individual (or local) explanations.
Some of the key features:
-
🏃 Fast recourse generation via
jax.jit,jax.vmap/jax.pmap. -
🚀 Accelerated over
cpu,gpu,tpu. -
🪓 Extensive recourse methods implemented for benchmarking.
-
👐 Customizable API to enable building entire modeling and interpreting pipeline.
Installation
The latest ReLax release can directly be installed from PyPI:
pip install jax-relax
or install directly from the repository:
pip install git+https://github.com/BirkhoffG/ReLax.git
To futher unleash the power of accelerators (i.e., GPU/TPU), we suggest
to first install this library via pip install jax-relax. Then, follow
steps in the official install
guidelines to install the
right version for GPU or TPU.
An Example of using ReLax
See Getting Started with ReLax.
Citing ReLax
To cite this repository:
@software{relax2023github,
author = {Hangzhi Guo and Xinchang Xiong and Amulya Yadav},
title = {{R}e{L}ax: Recourse Explanation Library in Jax},
url = {http://github.com/birkhoffg/ReLax},
version = {0.1.0},
year = {2023},
}
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file jax-relax-0.1.0.tar.gz.
File metadata
- Download URL: jax-relax-0.1.0.tar.gz
- Upload date:
- Size: 35.5 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/4.0.1 CPython/3.7.12
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
de3ca014526c63e7d3dad90c0de382020c76e3e6770e69e7a65136adddd77896
|
|
| MD5 |
fa6ca4a25a6e86e0876c143d306dbabf
|
|
| BLAKE2b-256 |
ed74dad6ed255b85e0f9432a5818b4c3915120b7362b965a11960f49c8ed56c5
|
File details
Details for the file jax_relax-0.1.0-py3-none-any.whl.
File metadata
- Download URL: jax_relax-0.1.0-py3-none-any.whl
- Upload date:
- Size: 42.8 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/4.0.1 CPython/3.7.12
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
dc7da04f8e18571f3b1ec274b3e196aac42d4fd2e3b5261b8be57f3a356aa14b
|
|
| MD5 |
b13e98935588918614d9e73d03c5315b
|
|
| BLAKE2b-256 |
d9db2527bdedf4c11329a53bc0f338af7c225736b3ca70ee15bf2d037bb3df44
|