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
- [WIP] [Graph] tests: benchdnn: in: graph: add f16 sample for GQA dropout bwd variant
- [WIP] xe: gated_mlp: improve performance of ukernel-based gmlp
- [GPU][NVL-P] Add acc mode restriction on atomic acc
- benchdnn: support SDPA primitive in verbose output
- benchdnn: inputs: graph: add torchbench sdpa training tests
- WIP: xe: conv: consolidate accumulator type setup
- cpu: x64: matmul: enable treat_as_plain for weights format
- Test PR. Upconvert fp8 weights to xf16 in Matmul in case of xf16 activations for 3.10
- cpu: x64: matmul: Enable int8 grouped quantization
- cpu: aarch64: enable binary op and binary post-ops on ASIMD
- Docs
- C++ not yet supported