ver4a/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

llama/ggml: add LLM training support (#10544)

Browse source 
* llama/ggml: add LLM training support

more compact progress bar

llama_save_model_to_file

llama_opt_param_filter

ggml_graph_dup force_grads

refactor ggml_opt, fix test-opt

* remove logits_all

* refactor CUDA implementation for ACC

* reset graph at beginning of opt period
This commit is contained in:
Johannes Gäßler 2025-05-12 14:44:49 +02:00 • committed by GitHub
parent 064cc596ac
commit 10d2af0eaa
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
31 changed files with 1415 additions and 359 deletions
Download patch file Download diff file Expand all files Collapse all files

4
src/llama-quant.cpp
View file

@ -519,7 +519,7 @@ static void llama_model_quantize_impl(const std::string & fname_inp, const std::
nthread = std::thread::hardware_concurrency();
}
// mmap consistently increases speed Linux, and also increases speed on Windows with
// mmap consistently increases speed on Linux, and also increases speed on Windows with
// hot cache. It may cause a slowdown on macOS, possibly related to free memory.
#if defined(__linux__) || defined(_WIN32)
constexpr bool use_mmap = true;
@ -529,7 +529,7 @@ static void llama_model_quantize_impl(const std::string & fname_inp, const std::
llama_model_kv_override * kv_overrides = nullptr;
if (params->kv_overrides) {
auto v = (std::vector<llama_model_kv_override>*)params->kv_overrides;
auto * v = (std::vector<llama_model_kv_override>*)params->kv_overrides;
kv_overrides = v->data();
}