onnxruntime
4afdc199
- ROCm optimized layernorm for MI100 (#7682)
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Commit
View On
GitHub
Commit
4 years ago
ROCm optimized layernorm for MI100 (#7682) * layernorm optimizations * Changed HIP flag from HIP_VERSION to __HIP_PLATFORM_HCC__
References
#7682 - ROCm optimized layernorm for MI100
Author
amathews-amd
Parents
d90a99aa
Loading