ROCm / Tensile

Stretching GPU performance for GEMMs and tensor contractions.

gemm blas dnn neural-networks machine-learning tensors python opencl hip radeon auto-tuning amd gpu-computing gpu-acceleration gpu matrix-multiplication tensor-contraction assembly

Tensile is a tool for creating benchmark-driven backend libraries for GEMMs, GEMM-like problems (such as batched GEMM), and general N-dimensional tensor contractions on a GPU. The Tensile library is mainly used as backend library to rocBLAS. Tensile acts as the performance backbone for a wide variety of 'compute' applications running on AMD GPUs.

See Tensile Wiki for documentation.

About

Stretching GPU performance for GEMMs and tensor contractions.

gemm blas dnn neural-networks machine-learning tensors python opencl hip radeon auto-tuning amd gpu-computing gpu-acceleration gpu matrix-multiplication tensor-contraction assembly

MIT License

Languages

Language:Python 50.2%Language:C++ 30.2%Language:Assembly 15.4%Language:TeX 1.4%Language:CMake 1.1%Language:Shell 1.1%Language:Groovy 0.4%Language:Makefile 0.1%Language:Awk 0.0%Language:Dockerfile 0.0%