root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-12-25 10:54:36 +00:00
Code Issues Actions 9 Packages Projects Releases Wiki Activity

train : move number of gpu layers argument parsing to common/train.cpp (#4074)

Browse Source 
- introduces help entry for the argument
 - cuts '--gpu-layers' form in order to simplify usage and documentation.

Signed-off-by: Jiri Podivin <jpodivin@gmail.com>
Co-authored-by: Jiri Podivin <jpodivin@redhat.com>
This commit is contained in:
Jiří Podivín 2023-11-17 16:19:16 +01:00 committed by GitHub
parent e85bb1a8e7
commit ba4cf5c0bf
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
2 changed files with 12 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
common/train.cpp
View File

@ -1136,6 +1136,7 @@ void print_common_train_usage(int /*argc*/, char ** /*argv*/, const struct train
fprintf(stderr, " --adam-beta2 N AdamW beta2 in interval [0,1). How much to smooth the second moment of gradients. (default %f)\n", params->adam_beta2); fprintf(stderr, " --adam-beta2 N AdamW beta2 in interval [0,1). How much to smooth the second moment of gradients. (default %f)\n", params->adam_beta2);
fprintf(stderr, " --adam-gclip N AdamW gradient clipping. Disabled when zero. (default %f)\n", params->adam_gclip); fprintf(stderr, " --adam-gclip N AdamW gradient clipping. Disabled when zero. (default %f)\n", params->adam_gclip);
fprintf(stderr, " --adam-epsf N AdamW epsilon for convergence test. Disabled when <= zero. (default %f)\n", params->adam_eps_f); fprintf(stderr, " --adam-epsf N AdamW epsilon for convergence test. Disabled when <= zero. (default %f)\n", params->adam_eps_f);
fprintf(stderr, " -ngl N, --n-gpu-layers N Number of model layers to offload to GPU (default %d)", params->n_gpu_layers);
fprintf(stderr, "\n"); fprintf(stderr, "\n");
} }
@ -1355,6 +1356,17 @@ bool consume_common_train_arg(
return true; return true;
} }
params->adam_gclip = std::stof(argv[i]); params->adam_gclip = std::stof(argv[i]);
} else if (arg == "-ngl" || arg == "--n-gpu-layers") {
if (++i >= argc) {
*invalid_param = true;
return true;
}
#ifdef LLAMA_SUPPORTS_GPU_OFFLOAD
params->n_gpu_layers = std::stoi(argv[i]);
#else
fprintf(stderr, "warning: not compiled with GPU offload support, --n-gpu-layers option will be ignored\n");
fprintf(stderr, "warning: see main README.md for information on enabling GPU BLAS support\n");
#endif
} else if (arg == "-h" || arg == "--help") { } else if (arg == "-h" || arg == "--help") {
params->print_usage = true; params->print_usage = true;
return true; return true;

11
examples/finetune/finetune.cpp
View File

@ -1460,17 +1460,6 @@ static bool train_params_parse(int argc, char ** argv, struct train_params * par
} }
params->n_rank_w3 = std::stoi(argv[i]); params->n_rank_w3 = std::stoi(argv[i]);
params->custom_n_rank_w3 = true; params->custom_n_rank_w3 = true;
} else if (arg == "--gpu-layers" || arg == "-ngl" || arg == "--n-gpu-layers") {
if (++i >= argc) {
invalid_param = true;
break;
}
#ifdef LLAMA_SUPPORTS_GPU_OFFLOAD
params->common.n_gpu_layers = std::stoi(argv[i]);
#else
fprintf(stderr, "warning: not compiled with GPU offload support, --n-gpu-layers option will be ignored\n");
fprintf(stderr, "warning: see main README.md for information on enabling GPU BLAS support\n");
#endif
} else { } else {
fprintf(stderr, "error: unknown argument: %s\n", arg.c_str()); fprintf(stderr, "error: unknown argument: %s\n", arg.c_str());
train_print_usage(argc, argv, &default_params); train_print_usage(argc, argv, &default_params);