onnxruntime
3bc9efc7
- [ORTModule] Adjust Attention Patterns for Efficient Attention ATen Fallback (#18471)
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Commit
View On
GitHub
Commit
2 years ago
[ORTModule] Adjust Attention Patterns for Efficient Attention ATen Fallback (#18471) Adjust attention patterns to match latest Whisper+exporter. Also add some condition check and add docs.
References
#18471 - [ORTModule] Adjust Attention Patterns for Efficient Attention ATen Fallback
Author
Lafi7e
Parents
7c573054
Loading