]> git.djapps.eu Git - pkg/ggml/sources/ggml/commit
ggml : sync with llama.cpp
authorGeorgi Gerganov <redacted>
Mon, 10 Apr 2023 16:36:06 +0000 (19:36 +0300)
committerGeorgi Gerganov <redacted>
Mon, 10 Apr 2023 17:40:36 +0000 (20:40 +0300)
commit3ac8072331649cd6388f52a076f530e607980d18
treec20f284e8ca870f537e93221c4504bdd07348ee2
parent42cbb07a6a415d041a76138dc1d47f3d6697e501
ggml : sync with llama.cpp

- int64_t number of elements
- remove mlock
- expose quantization functions
- expose ggml_object
- add ggml_view_3d()
- multi-thread ggml_rope()
- fix ggml_cpy()
- add ggml_init_params.no_alloc
- fix ggml_mul_mat() backward
18 files changed:
examples/gpt-2/main.cpp
examples/gpt-2/quantize.cpp
examples/gpt-j/main.cpp
examples/gpt-j/quantize.cpp
examples/whisper/main.cpp
examples/whisper/quantize.cpp
examples/whisper/whisper.cpp
examples/whisper/whisper.h
include/ggml/ggml.h
src/ggml.c
tests/test-blas0.c
tests/test-grad0.c
tests/test-mul-mat0.c
tests/test-mul-mat2.c
tests/test0.c
tests/test1.c
tests/test2.c
tests/test3.c