FairScale: A PyTorch library for large-scale and high-performance training.
Project Links
Meta
Author: Foundational AI Research @ Meta AI
Requires Python: >=3.8
Classifiers
Programming Language
- Python :: 3.8
- Python :: 3.9
- Python :: 3.10
License
- OSI Approved :: BSD License
Topic
- Scientific/Engineering :: Artificial Intelligence
Operating System
- OS Independent
FairScale is a PyTorch extension library for high performance and large scale training on one or multiple machines/nodes. This library extends basic PyTorch capabilities while adding new experimental ones.
0.4.13
Dec 11, 2022
0.4.12
Oct 05, 2022
0.4.11
Sep 30, 2022
0.4.10
Sep 23, 2022
0.4.9
Sep 07, 2022
0.4.8
Jul 26, 2022
0.4.7
Jul 26, 2022
0.4.6
Mar 09, 2022
0.4.5
Jan 14, 2022
0.4.4
Dec 21, 2021
0.4.3
Nov 18, 2021
0.4.2
Nov 08, 2021
0.4.1
Oct 08, 2021
0.4.0
Aug 01, 2021
0.3.9
Jul 27, 2021
0.3.8
Jul 12, 2021
0.3.7
May 18, 2021
0.3.6
Apr 26, 2021
0.3.5
Apr 19, 2021
0.3.4
Apr 13, 2021
0.3.3
Apr 02, 2021
0.3.2
Mar 18, 2021
0.3.1
Mar 09, 2021
0.3.0
Feb 23, 2021
0.1.7
Feb 19, 2021
0.1.6
Feb 11, 2021
0.1.5
Feb 03, 2021
0.1.4
Jan 07, 2021
0.1.3
Jan 05, 2021
0.1.1
Dec 02, 2020
0.1.0
Dec 01, 2020
0.0.3
Oct 15, 2020
0.0.2
Aug 28, 2020
0.0.1
Jul 31, 2020
0.0.0
Jul 31, 2020
Files in release
No dependencies