mirror of
https://github.com/ggml-org/llama.cpp.git
synced 2025-08-17 13:40:55 -04:00
c959f462a0
* CUDA: add conv_2d_transpose * remove direct include of cuda_fp16 * Review: add brackets for readability, remove ggml_set_param and add asserts
5 lines
157 B
Plaintext
5 lines
157 B
Plaintext
#include "common.cuh"
|
|
|
|
#define CUDA_CONV2D_TRANSPOSE_BLOCK_SIZE 256
|
|
void ggml_cuda_conv_2d_transpose_p0(ggml_backend_cuda_context & ctx, ggml_tensor * dst);
|