CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-10-30 08:42:00 +00:00
Code Issues Packages Projects Releases Wiki Activity

llama : uniform variable names + struct init

Browse Source
This commit is contained in:
Georgi Gerganov
2023-07-05 23:15:54 +03:00
parent 4fe95c6985
commit 26cc1bd7a2
3 changed files with 29 additions and 29 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
examples/server/server.cpp
View File

@@ -915,14 +915,14 @@ static json format_timings(llama_server_context &llama)
return json{
{"prompt_n", timings.n_eval},
{"prompt_ms", timings.prompt_eval_time_ms},
{"prompt_per_token_ms", timings.prompt_eval_time_ms / timings.n_p_eval},
{"prompt_per_second", 1e3 / timings.prompt_eval_time_ms * timings.n_p_eval},
{"prompt_ms", timings.t_p_eval_ms},
{"prompt_per_token_ms", timings.t_p_eval_ms / timings.n_p_eval},
{"prompt_per_second", 1e3 / timings.t_p_eval_ms * timings.n_p_eval},
{"predicted_n", timings.n_eval},
{"predicted_ms", timings.eval_time_ms},
{"predicted_per_token_ms", timings.eval_time_ms / timings.n_eval},
{"predicted_per_second", 1e3 / timings.eval_time_ms * timings.n_eval},
{"predicted_ms", timings.t_eval_ms},
{"predicted_per_token_ms", timings.t_eval_ms / timings.n_eval},
{"predicted_per_second", 1e3 / timings.t_eval_ms * timings.n_eval},
};
}

36
llama.cpp
View File

@@ -3480,34 +3480,34 @@ llama_token llama_token_nl() {
return 13;
}
llama_timings llama_get_timings(struct llama_context * ctx) {
llama_timings timings;
struct llama_timings llama_get_timings(struct llama_context * ctx) {
struct llama_timings result = {
/*.t_start_ms =*/ 1e-3 * ctx->t_start_us,
/*.t_end_ms =*/ 1.00 * ggml_time_ms(),
/*.t_load_ms =*/ 1e-3 * ctx->t_load_us,
/*.t_sample_ms =*/ 1e-3 * ctx->t_sample_us,
/*.t_p_eval_ms =*/ 1e-3 * ctx->t_p_eval_us,
/*.t_eval_ms =*/ 1e-3 * ctx->t_eval_us,
timings.t_end_ms = ggml_time_ms();
timings.t_start_ms = 1e-3 * ctx->t_start_us;
timings.load_time_ms = 1e-3 * ctx->t_load_us;
timings.sample_time_ms = 1e-3 * ctx->t_sample_us;
timings.prompt_eval_time_ms = 1e-3 * ctx->t_p_eval_us;
timings.eval_time_ms = 1e-3 * ctx->t_eval_us;
/*.n_sample =*/ std::max(1, ctx->n_sample),
/*.n_p_eval =*/ std::max(1, ctx->n_p_eval),
/*.n_eval =*/ std::max(1, ctx->n_eval),
};
timings.n_sample = std::max(1, ctx->n_sample);
timings.n_p_eval = std::max(1, ctx->n_p_eval);
timings.n_eval = std::max(1, ctx->n_eval);
return timings;
return result;
}
void llama_print_timings(struct llama_context * ctx) {
llama_timings timings = llama_get_timings(ctx);
const llama_timings timings = llama_get_timings(ctx);
fprintf(stderr, "\n");
fprintf(stderr, "%s: load time = %8.2f ms\n", __func__, timings.load_time_ms);
fprintf(stderr, "%s: load time = %8.2f ms\n", __func__, timings.t_load_ms);
fprintf(stderr, "%s: sample time = %8.2f ms / %5d runs (%8.2f ms per token, %8.2f tokens per second)\n",
__func__, timings.sample_time_ms, timings.n_sample, timings.sample_time_ms / timings.n_sample, 1e3 / timings.sample_time_ms * timings.n_sample);
__func__, timings.t_sample_ms, timings.n_sample, timings.t_sample_ms / timings.n_sample, 1e3 / timings.t_sample_ms * timings.n_sample);
fprintf(stderr, "%s: prompt eval time = %8.2f ms / %5d tokens (%8.2f ms per token, %8.2f tokens per second)\n",
__func__, timings.prompt_eval_time_ms, timings.n_p_eval, timings.prompt_eval_time_ms / timings.n_p_eval, 1e3 / timings.prompt_eval_time_ms * timings.n_p_eval);
__func__, timings.t_p_eval_ms, timings.n_p_eval, timings.t_p_eval_ms / timings.n_p_eval, 1e3 / timings.t_p_eval_ms * timings.n_p_eval);
fprintf(stderr, "%s: eval time = %8.2f ms / %5d runs (%8.2f ms per token, %8.2f tokens per second)\n",
__func__, timings.eval_time_ms, timings.n_eval, timings.eval_time_ms / timings.n_eval, 1e3 / timings.eval_time_ms * timings.n_eval);
__func__, timings.t_eval_ms, timings.n_eval, timings.t_eval_ms / timings.n_eval, 1e3 / timings.t_eval_ms * timings.n_eval);
fprintf(stderr, "%s: total time = %8.2f ms\n", __func__, (timings.t_end_ms - timings.t_start_ms));
}

10
llama.h
View File

@@ -138,10 +138,10 @@ extern "C" {
struct llama_timings {
double t_start_ms;
double t_end_ms;
double load_time_ms;
double sample_time_ms;
double prompt_eval_time_ms;
double eval_time_ms;
double t_load_ms;
double t_sample_ms;
double t_p_eval_ms;
double t_eval_ms;
int32_t n_sample;
int32_t n_p_eval;
@@ -345,7 +345,7 @@ extern "C" {
LLAMA_API llama_token llama_sample_token(struct llama_context * ctx, llama_token_data_array * candidates);
// Performance information
LLAMA_API llama_timings llama_get_timings(struct llama_context * ctx);
LLAMA_API struct llama_timings llama_get_timings(struct llama_context * ctx);
LLAMA_API void llama_print_timings(struct llama_context * ctx);
LLAMA_API void llama_reset_timings(struct llama_context * ctx);