Post
355
Building high-performance, reproducible kernels forĀ AMD ROCmĀ just got a lot easier.
I've put together a guide on building, testing, and sharingĀ ROCm-compatible kernelsĀ using the Hugging FaceĀ
Learn how to:
- Use Nix for reproducible builds
- Integrate kernels as native PyTorch operators
- Share your kernels on the Hub for anyone to use withĀ
We use the š award-winning RadeonFlow GEMM kernel as a practical example.
š Check out the full guide here : https://huggingface.co/blog/build-rocm-kernels
I've put together a guide on building, testing, and sharingĀ ROCm-compatible kernelsĀ using the Hugging FaceĀ
kernel-builder and kernelsĀ libraries; so you can focus onĀ optimizing performanceĀ rather than spending time on setup.Learn how to:
- Use Nix for reproducible builds
- Integrate kernels as native PyTorch operators
- Share your kernels on the Hub for anyone to use withĀ
kernels.get_kernel()We use the š award-winning RadeonFlow GEMM kernel as a practical example.
š Check out the full guide here : https://huggingface.co/blog/build-rocm-kernels
