tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-08-31 12:22:58 -04:00
Code Issues Packages Projects Releases Wiki Activity

ggml : ggml_set_rows support broadcast

Browse Source
This commit is contained in:
Georgi Gerganov
2025-06-22 10:28:07 +03:00
parent 313a444b22
commit df71c803b4
3 changed files with 39 additions and 15 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
ggml/include/ggml.h
View File

@@ -1379,6 +1379,15 @@ extern "C" {
struct ggml_tensor * b, // row indices
struct ggml_tensor * c); // data for ggml_get_rows, only used for its shape
// a TD [n_embd, ne1, ne2, ne3]
// b TS [n_embd, n_rows, ne02, ne03] | ne02 == ne2, ne03 == ne3
// c I64 [n_rows, ne11, ne12, 1] | c[i] in [0, ne1)
//
// broadcast:
// ne2 % ne11 == 0
// ne3 % ne12 == 0
//
// return view(a)
GGML_API struct ggml_tensor * ggml_set_rows(
struct ggml_context * ctx,
struct ggml_tensor * a, // destination