NVIDIA / TransformerEngine

A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit floating point (FP8) precision on Hopper and Ada GPUs, to provide better performance with lower memory utilization in both training and inference.
https://docs.nvidia.com/deeplearning/transformer-engine/user-guide/index.html
Apache License 2.0
2k stars 333 forks source link

`intra_domain_ranks` is not defined in one of the execution paths. #1306

Closed erhoo82 closed 3 weeks ago

erhoo82 commented 3 weeks ago

intra_domain_ranks at https://github.com/NVIDIA/TransformerEngine/blob/main/transformer_engine/pytorch/module/base.py#L257C68-L257C86 is not defined when if num_domains > 1.

denera commented 3 weeks ago

Thanks! This was internally reported earlier today as well and PR #1305 should fix it.