Fix layernorm implementation
Showing
- Makefile 35 additions, 33 deletionsMakefile
- dnn_kernels/benchmarks/layernorm-fp64-48-48.c 5 additions, 0 deletionsdnn_kernels/benchmarks/layernorm-fp64-48-48.c
- dnn_kernels/benchmarks/layernorm-fp64-sdma-48-48.c 5 additions, 0 deletionsdnn_kernels/benchmarks/layernorm-fp64-sdma-48-48.c
- dnn_kernels/benchmarks/layernorm-fp64-sdma-ssr-48-48.c 5 additions, 0 deletionsdnn_kernels/benchmarks/layernorm-fp64-sdma-ssr-48-48.c
- dnn_kernels/benchmarks/layernorm-fp64-sdma-ssr-frep-48-48.c 2 additions, 2 deletionsdnn_kernels/benchmarks/layernorm-fp64-sdma-ssr-frep-48-48.c
- dnn_kernels/benchmarks/layernorm-fp64-sdma-ssr-frep-omp-48-48.c 2 additions, 2 deletions...rnels/benchmarks/layernorm-fp64-sdma-ssr-frep-omp-48-48.c
- dnn_kernels/benchmarks/layernorm-raw-fp64-sdma-ssr-frep-48-48.c 2 additions, 2 deletions...rnels/benchmarks/layernorm-raw-fp64-sdma-ssr-frep-48-48.c
- dnn_kernels/benchmarks/templates/layernorm_raw.tpl.c 12 additions, 12 deletionsdnn_kernels/benchmarks/templates/layernorm_raw.tpl.c
- dnn_kernels/src/layernorm.c 65 additions, 112 deletionsdnn_kernels/src/layernorm.c
Please register or sign in to comment