onnxruntime
Optimize cuComputePartGradGammaBeta kernel for MI100
#10475

Merged

Optimize cuComputePartGradGammaBeta kernel for MI100 #10475

weixingzhang merged 3 commits into microsoft:master from ROCm:rocm_layernorm_optim

Optimize cuComputePartGradGammaBeta kernel for MI100

8ed46ca5

jeffdaily requested changes on 2022-02-08

Update orttraining/orttraining/training_ops/cuda/nn/layer_norm.cc

63323d08

Update orttraining/orttraining/training_ops/cuda/nn/layer_norm.cc

edc025fa

weixingzhang approved these changes on 2022-02-09

weixingzhang merged c9fbd0b1 into master 4 years ago

Reviewers

weixingzhang

jeffdaily

Assignees

No one assigned

Labels

None yet

Milestone

No milestone