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

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
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-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, " -ngl N, --n-gpu-layers N Number of model layers to offload to GPU (default %d)", params->n_gpu_layers);
fprintf(stderr, "\n");
}
@ -1355,6 +1356,17 @@ bool consume_common_train_arg(
return true;
}
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") {
params->print_usage = true;
return true;