tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-06-27 12:05:03 +00:00
Code Issues Packages Projects Releases Wiki Activity

sampling : refactor init to use llama_sampling_params (#3696)

Browse Source 
* sampling : refactor init to use llama_sampling_params

* llama : combine repetition, frequency and presence penalties in 1 call

* examples : remove embd-input and gptneox-wip

* sampling : rename penalty params + reduce size of "prev" vector

* sampling : add llama_sampling_print helper

* sampling : hide prev behind API and apply #3661

ggml-ci
This commit is contained in:
Georgi Gerganov
2023-10-20 21:07:23 +03:00
committed by GitHub
parent 8cf19d60dc
commit d1031cf49c
30 changed files with 365 additions and 4502 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
examples/parallel/parallel.cpp
View File

@ -157,7 +157,7 @@ int main(int argc, char ** argv) {
for (size_t i = 0; i < clients.size(); ++i) {
auto & client = clients[i];
client.id = i;
client.ctx_sampling = llama_sampling_init(params);
client.ctx_sampling = llama_sampling_init(params.sparams);
}
std::vector<llama_token> tokens_system;
@ -330,7 +330,7 @@ int main(int argc, char ** argv) {
const llama_token id = llama_sampling_sample(client.ctx_sampling, ctx, NULL, client.i_batch - i);
llama_sampling_accept(client.ctx_sampling, ctx, id);
llama_sampling_accept(client.ctx_sampling, ctx, id, true);
if (client.n_decoded == 1) {
// start measuring generation time after the first token to make sure all concurrent clients