Skip to content

replace GPU 1./sqrt with rsqrt#1741

Merged
wanghan-iapcm merged 3 commits into
deepmodeling:develfrom
njzjz:cuda-rsqrt
Jun 7, 2022
Merged

replace GPU 1./sqrt with rsqrt#1741
wanghan-iapcm merged 3 commits into
deepmodeling:develfrom
njzjz:cuda-rsqrt

apply the same opt for ROCM

486fb8c
Select commit
Loading
Failed to load commit list.

Workflow runs completed with no jobs