Mirrors/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-12-24 13:28:50 +01:00
Code Issues Packages Projects Releases Wiki Activity

llama-bench : minor fixes (#2695)

Browse Source
This commit is contained in:
slaren 2023-08-22 09:56:03 +02:00 committed by GitHub
parent 1e3bc523d8
commit 8e4364f2af
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 5 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
examples/llama-bench/llama-bench.cpp
View File

@ -148,7 +148,7 @@ struct cmd_params {
}; };
static const cmd_params cmd_params_defaults = { static const cmd_params cmd_params_defaults = {
/* model */ {"models/7B/ggml-model-q4_0.bin"}, /* model */ {"models/7B/ggml-model-q4_0.gguf"},
/* n_prompt */ {512}, /* n_prompt */ {512},
/* n_gen */ {128}, /* n_gen */ {128},
/* n_batch */ {512}, /* n_batch */ {512},
@ -179,12 +179,12 @@ static void print_usage(int /* argc */, char ** argv) {
fprintf(stdout, " -mg i, --main-gpu <n> (default: %s)\n", join(cmd_params_defaults.main_gpu, ",").c_str()); fprintf(stdout, " -mg i, --main-gpu <n> (default: %s)\n", join(cmd_params_defaults.main_gpu, ",").c_str());
fprintf(stdout, " -lv, --low-vram <0|1> (default: %s)\n", join(cmd_params_defaults.low_vram, ",").c_str()); fprintf(stdout, " -lv, --low-vram <0|1> (default: %s)\n", join(cmd_params_defaults.low_vram, ",").c_str());
fprintf(stdout, " -mmq, --mul-mat-q <0|1> (default: %s)\n", join(cmd_params_defaults.mul_mat_q, ",").c_str()); fprintf(stdout, " -mmq, --mul-mat-q <0|1> (default: %s)\n", join(cmd_params_defaults.mul_mat_q, ",").c_str());
fprintf(stdout, " -ts, --tensor_split <ts> \n"); fprintf(stdout, " -ts, --tensor_split <ts0/ts1/..> \n");
fprintf(stdout, " -r, --repetitions <n> (default: %d)\n", cmd_params_defaults.reps); fprintf(stdout, " -r, --repetitions <n> (default: %d)\n", cmd_params_defaults.reps);
fprintf(stdout, " -o, --output <csv|json|md|sql> (default: %s)\n", cmd_params_defaults.output_format == CSV ? "csv" : cmd_params_defaults.output_format == JSON ? "json" : "md"); fprintf(stdout, " -o, --output <csv|json|md|sql> (default: %s)\n", cmd_params_defaults.output_format == CSV ? "csv" : cmd_params_defaults.output_format == JSON ? "json" : cmd_params_defaults.output_format == MARKDOWN ? "md" : "sql");
fprintf(stdout, " -v, --verbose (default: %s)\n", cmd_params_defaults.verbose ? "1" : "0"); fprintf(stdout, " -v, --verbose (default: %s)\n", cmd_params_defaults.verbose ? "1" : "0");
fprintf(stdout, "\n"); fprintf(stdout, "\n");
fprintf(stdout, "Multiple values can be given for each parameter by separating them with ',' or by repeating the parameter.\n"); fprintf(stdout, "Multiple values can be given for each parameter by separating them with ',' or by specifying the parameter multiple times.\n");
} }
@ -728,7 +728,7 @@ struct markdown_printer : public printer {
if (!is_cpu_backend) { if (!is_cpu_backend) {
fields.push_back("n_gpu_layers"); fields.push_back("n_gpu_layers");
} }
if (params.n_batch.size() > 1 || params.n_threads != cmd_params_defaults.n_threads || is_cpu_backend) { if (params.n_threads.size() > 1 || params.n_threads != cmd_params_defaults.n_threads || is_cpu_backend) {
fields.push_back("n_threads"); fields.push_back("n_threads");
} }
if (params.n_batch.size() > 1 || params.n_batch != cmd_params_defaults.n_batch) { if (params.n_batch.size() > 1 || params.n_batch != cmd_params_defaults.n_batch) {