vrr/prima.cpp
2
0
Fork 0
mirror of https://github.com/Lizonghang/prima.cpp.git synced 2025-09-10 11:24:40 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

add llama_model_n_flops

Browse source
This commit is contained in:
Lizonghang 2024-11-20 19:40:27 +04:00
parent 10f6f92c7e
commit 477ecf2084
4 changed files with 445 additions and 107 deletions
Download patch file Download diff file Expand all files Collapse all files

16
include/llama.h
View file

@ -410,7 +410,13 @@ extern "C" {
// Call once at the start of the program
LLAMA_API void llama_backend_init(void);
LLAMA_API void llama_profile_device (struct device_info * dev_info, struct llama_model * model, const char * test_file, int n_threads);
LLAMA_API void llama_profile_device(
struct device_info * dev_info,
struct llama_model * model,
struct llama_model_loader * ml,
const char * test_file,
int n_threads);
LLAMA_API ggml_backend_buffer_type_t llama_dev_buffer_type(struct llama_model * model, int device);
//optional:
@ -518,6 +524,14 @@ extern "C" {
// Returns the total number of parameters in the model
LLAMA_API uint64_t llama_model_n_params(const struct llama_model * model);
// Return the total number of float operations in the model
LLAMA_API void llama_model_n_flops(
struct llama_model * model,
struct llama_model_loader * ml,
struct flops_info * ffo,
const int64_t n_input,
const int64_t n_history);
// Get a llama model tensor
LLAMA_API struct ggml_tensor * llama_get_model_tensor(struct llama_model * model, const char * name);