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
- CPU: RV64: enable f16 destination for BRGEMM inner product
- CPU: RV64: use widening accumulation for f16 layer normalization
- CPU: RV64: widen f16 max pooling vectors for VLEN=256 nspc
- CPU: RV64: support f32 bias in f16 and bf16 matmul JIT
- CPU: RV64: hoist xf16 softmax exp coefficients on VLEN >= 256
- ngen: pull down from upstream
- 1x1 convolution backward_weights writes past the rtus workspace (nhwc, IC < ic_block)
- [rls-v3.14][CPU] Asynchronous verbose mode for async threadpool
- rfcs: extend GPU JIT GEMM kernel selection
- xe: ggemm: Add support for 2d quant scales and zp. Undo A_copies in ukernel selector
- Docs
- C++ not yet supported