root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-12-25 19:04:35 +00:00
Code Issues Actions 7 Packages Projects Releases Wiki Activity

remove candidates_decoded

Browse Source
This commit is contained in:
marcus 2023-12-04 14:11:45 -08:00
parent eb9d1fcd7d
commit 3773328080
1 changed files with 2 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
llama.cpp
View File

@ -7356,8 +7356,6 @@ void llama_sample_grammar(struct llama_context * ctx, llama_token_data_array * c
const llama_token eos = llama_token_eos(&ctx->model);
std::vector<std::pair<std::vector<uint32_t>, llama_partial_utf8>> candidates_decoded;
candidates_decoded.reserve(candidates->size);
std::vector<llama_grammar_candidate> candidates_grammar;
candidates_grammar.reserve(candidates->size);
@ -7371,8 +7369,8 @@ void llama_sample_grammar(struct llama_context * ctx, llama_token_data_array * c
} else if (piece.empty() || piece[0] == 0) {
candidates->data[i].logit = -INFINITY;
} else {
candidates_decoded.push_back(decode_utf8(piece, grammar->partial_utf8));
candidates_grammar.push_back({ i, candidates_decoded.back().first.data(), candidates_decoded.back().second });
std::pair<std::vector<uint32_t>, llama_partial_utf8> decoded = decode_utf8(piece, grammar->partial_utf8);
candidates_grammar.push_back({ i, decoded.first.data(), decoded.second });
}
}