onednn
https://github.com/oneapi-src/onednn
C++
oneAPI Deep Neural Network Library (oneDNN)
Triage Issues!
When you volunteer to triage issues, you'll receive an email each day with a link to an open issue that needs help in this project. You'll also receive instructions on how to triage issues.
Triage Docs!
Receive a documented method or class from your favorite GitHub repos in your inbox every day. If you're really pro, receive undocumented methods or classes and supercharge your commit history.
C++ not yet supported2 Subscribers
Add a CodeTriage badge to onednn
Help out
- Issues
- common: in-place binary post-op takes the destination descriptor
- xe: matmul: gemm: keep per-row scales out of the batch-merging reshape
- cpu: matmul: gemm: reject weight scales over K
- xe: sdpa: enable fused backward on xe_hpg
- gpu: jit:gemm: s32 destination with scales and large K writes float bit patterns
- gpu: jit:gemm: u4 weights with a src zero-point and a div post-op differ by one bf16 step
- xe: sdpa: retune the xe_hpg head-size-64 thin-query rows
- github: workflows: bump github/codeql-action/analyze from 4.39.1 to 4.39.3
- github: workflows: bump github/codeql-action/init from 4.39.1 to 4.39.3
- Reduce creation time of IR-based kernels
- Docs
- C++ not yet supported