Shadowgraph
About
Sign in
← Papers

System Optimizations for Enabling Training of Extreme Long Sequence Transformer Models

ACM:3662158.3662806 · 2 repos reference this paper in their README

deepspeedai/DeepSpeed

43,096

·

+32

·

Python

DeepSpeed is a deep learning optimization library that makes distributed training and inference…

microsoft/DeepSpeed

43,096

·

+32

·

Python

DeepSpeed is a deep learning optimization library that makes distributed training and inference…