tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-06-29 04:35:05 +00:00
Code Issues Packages Projects Releases Wiki Activity

llama : refactor samplers internal implementation (#9370)

Browse Source
This commit is contained in:
slaren
2024-09-08 15:52:07 +02:00
committed by GitHub
parent 2a358fb0c4
commit 19f4a7b296
4 changed files with 881 additions and 725 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
src/llama-sampling.h
View File

@ -23,16 +23,6 @@ struct llama_sampler_chain {
mutable int32_t n_sample;
};
using llama_token_cnt = std::unordered_map<llama_token, int>;
// TODO: tmp exposed until test-sampling is fixed
void llama_sampler_penalties_impl(
llama_token_data_array * cur_p,
const llama_token_cnt & token_count,
float penalty_repeat,
float penalty_freq,
float penalty_present);
struct llama_sampler * llama_sampler_init_grammar_impl(
const struct llama_vocab & vocab,
const char * grammar_str,