root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-11-11 13:30:35 +00:00
Code Issues Actions 11 Packages Projects Releases Wiki Activity

llama : use reserve/emplace_back in sampler_sample (#9534)

Browse Source 
This commit updates the llama_sampler_sample function to use reserve and
emplace_back for the vector of llama_token_data structs.

The motivation for this change is to avoid the creation of n_vocab
default-constructed llama_token_data structs which are then
immediately overwritten.
This commit is contained in:
Daniel Bevenius 2024-09-18 13:42:36 +02:00 committed by GitHub
parent 8a308354f6
commit 6443ddd985
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 3 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
src/llama-sampling.cpp
View File

@ -236,9 +236,10 @@ llama_token llama_sampler_sample(struct llama_sampler * smpl, struct llama_conte
const int n_vocab = llama_n_vocab(llama_get_model(ctx)); const int n_vocab = llama_n_vocab(llama_get_model(ctx));
// TODO: do not allocate each time // TODO: do not allocate each time
std::vector<llama_token_data> cur(n_vocab); std::vector<llama_token_data> cur;
cur.reserve(n_vocab);
for (llama_token token_id = 0; token_id < n_vocab; token_id++) { for (llama_token token_id = 0; token_id < n_vocab; token_id++) {
cur[token_id] = llama_token_data{token_id, logits[token_id], 0.0f}; cur.emplace_back(llama_token_data{token_id, logits[token_id], 0.0f});
} }
llama_token_data_array cur_p = { llama_token_data_array cur_p = {