root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-11-14 23:09:53 +00:00
Code Issues Actions 14 Packages Projects Releases Wiki Activity

speculative : fix handling of some input params (#9963)

Browse Source 
* speculative : fix batch sizes at initialization

ggml-ci

* speculative : handle params.n_predict == -1

* speculative : limit batch size to llama_n_batch
This commit is contained in:
Georgi Gerganov 2024-10-21 09:37:12 +03:00 committed by GitHub
parent 1db8c84fc6
commit bc21975084
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 8 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
examples/speculative/speculative.cpp
View File

@ -39,6 +39,11 @@ int main(int argc, char ** argv) {
return 1;
}
if (params.n_predict < -1) {
LOG_ERR("%s: --n-predict must be >= -1\n", __func__);
return 1;
}
common_init();
if (params.model_draft.empty()) {
@ -190,8 +195,8 @@ int main(int argc, char ** argv) {
drafts[s].smpl = common_sampler_init(model_dft, params.sparams);
}
llama_batch batch_dft = llama_batch_init(params.n_ctx, 0, 1);
llama_batch batch_tgt = llama_batch_init(params.n_ctx, 0, n_seq_dft);
llama_batch batch_dft = llama_batch_init(llama_n_batch(ctx_dft), 0, 1);
llama_batch batch_tgt = llama_batch_init(llama_n_batch(ctx_tgt), 0, n_seq_dft);
const auto t_dec_start = ggml_time_us();
@ -441,7 +446,7 @@ int main(int argc, char ** argv) {
++n_past_dft;
}
if (n_predict > params.n_predict || has_eos) {
if ((params.n_predict >= 0 && n_predict > params.n_predict) || has_eos) {
break;
}