c1f1e28d29
* CUDA: add fast walsh-hadamard transform * review: add unrolls + change size_t -> int * warp size 64 --------- Co-authored-by: Johannes Gäßler <johannesg@5d6.de>
4 lines
124 B
Plaintext
4 lines
124 B
Plaintext
#include "common.cuh"
|
|
|
|
void ggml_cuda_op_fwht(ggml_backend_cuda_context & ctx, const ggml_tensor * src, ggml_tensor * dst);
|