piwheels

fp8-coat

https://arxiv.org/abs/2410.19313. COAT: Compressing Optimizer states and Activation for Memory-Efficient FP8 Training

Installation

In a virtualenv (see these instructions if you need to create one):

pip3 install fp8-coat

Dependencies

• wandb
• peft
• ipython
• transformers
• matplotlib
• sentencepiece
• torch
• triton
• gdown

Releases

Version	Released	Bullseye Python 3.9	Bookworm Python 3.11	Trixie Python 3.13	Files
1.0.11	2025-04-22
fp8_coat-1.0.11-py3-none-any.whl (137 KB)
1.0.10	2025-04-22
fp8_coat-1.0.10-py3-none-any.whl (137 KB)
1.0.9	2025-04-18
fp8_coat-1.0.9-py3-none-any.whl (137 KB)
1.0.8	2025-04-09
fp8_coat-1.0.8-py3-none-any.whl (137 KB)
1.0.7	2025-02-16
fp8_coat-1.0.7-py3-none-any.whl (121 KB)
1.0.6	2025-02-16
fp8_coat-1.0.6-py3-none-any.whl (8 KB)
1.0.5	2025-01-10
fp8_coat-1.0.5-py3-none-any.whl (121 KB)
1.0.4	2025-01-10
fp8_coat-1.0.4-py3-none-any.whl (116 KB)
1.0.3	2025-01-10
fp8_coat-1.0.3-py3-none-any.whl (116 KB)
1.0.2	2025-01-10
fp8_coat-1.0.2-py3-none-any.whl (14 KB)
1.0.0	2025-01-09
fp8_coat-1.0.0-py3-none-any.whl (112 KB)

Issues with this package?

• Search issues for this package
• Package or version missing? Open a new issue
• Something else? Open a new issue