vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

finetune : add -ngl parameter (#3762)

Browse source 
* Add '-ngl' support to finetune.cpp

* Add fprintf in ggml_cuda_op_add

When I tried CUDA offloading during finetuning following the readme, I got an assert here.
This probably isn't an important case because inference later gives a warning saying you should use f16 or f32 instead when using lora

* Add 'finetune.sh', which currently fails when using GPU

"error: operator (): Finetuning on tensors with type 'f16' is not yet supported"

* tweak finetune.sh

* Suppress some warnings in ggml.c

* Add f16 implementation to ggml_compute_forward_add_f16_f32

* Add an f16 case to ggml_add_cast_impl and llama_build_lora_finetune_graphs

* finetune.sh: Edit comments

* Add "add_f16_f32_f32_cuda"

* Tweak an error message

* finetune.sh: Add an optional LLAMA_MODEL_DIR variable

* finetune.sh: Add an optional LLAMA_TRAINING_DIR variable

* train : minor

* tabs to spaces

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
Co-authored-by: cebtenzzre <cebtenzzre@gmail.com>
This commit is contained in:
Andrew Godfrey 2023-11-01 04:49:04 -07:00 • committed by GitHub
parent f0e209324a
commit 73bdcb395e
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
8 changed files with 106 additions and 15 deletions
Download patch file Download diff file Expand all files Collapse all files

2
ggml-quants.c
View file

@ -716,6 +716,7 @@ void quantize_row_q8_0(const float * restrict x, void * restrict vy, int k) {
__riscv_vse8_v_i8m1(y[i].qs , vs, vl);
}
#else
UNUSED(nb);
// scalar
quantize_row_q8_0_reference(x, y, k);
#endif
@ -969,6 +970,7 @@ void quantize_row_q8_1(const float * restrict x, void * restrict vy, int k) {
y[i].s = sum*d;
}
#else
UNUSED(nb);
// scalar
quantize_row_q8_1_reference(x, y, k);
#endif