ikawrakow/ik_llama.cpp
1
0
Fork 0
mirror of https://github.com/ikawrakow/ik_llama.cpp.git synced 2026-04-20 22:49:31 +00:00
Code Issues Packages Projects Releases Wiki Activity

Server: refactor and rename functions (#1151)

Browse Source 
* Server: rename functions and refactor code

rename functions

refactor update slots

rename params_base

rename timings

* change

* Revert kv cache name changes

* Revert 2

* fix test build error

---------

Co-authored-by: firecoperana <firecoperana>
This commit is contained in:
firecoperana
2026-01-18 00:16:57 -06:00
committed by GitHub
parent 7024fdbc72
commit d71a3ec315
38 changed files with 532 additions and 528 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
examples/retrieval/retrieval.cpp
View File

@@ -76,7 +76,7 @@ static std::vector<chunk> chunk_file(const std::string & filename, int chunk_siz
static void batch_add_seq(llama_batch & batch, const std::vector<int32_t> & tokens, llama_seq_id seq_id) {
size_t n_tokens = tokens.size();
for (size_t i = 0; i < n_tokens; i++) {
llama_batch_add(batch, tokens[i], i, { seq_id }, true);
common_batch_add(batch, tokens[i], i, { seq_id }, true);
}
}
@@ -204,7 +204,7 @@ int main(int argc, char ** argv) {
fprintf(stderr, "%s: prompt %d: '%s'\n", __func__, i, chunks[i].textdata.c_str());
fprintf(stderr, "%s: number of tokens in prompt = %zu\n", __func__, chunks[i].tokens.size());
for (int j = 0; j < (int) chunks[i].tokens.size(); j++) {
fprintf(stderr, "%6d -> '%s'\n", chunks[i].tokens[j], llama_token_to_piece(ctx, chunks[i].tokens[j]).c_str());
fprintf(stderr, "%6d -> '%s'\n", chunks[i].tokens[j], common_token_to_piece(ctx, chunks[i].tokens[j]).c_str());
}
fprintf(stderr, "\n\n");
}
@@ -232,7 +232,7 @@ int main(int argc, char ** argv) {
if (batch.n_tokens + n_toks > n_batch) {
float * out = emb + p * n_embd;
batch_decode(ctx, batch, out, s, n_embd);
llama_batch_clear(batch);
common_batch_clear(batch);
p += s;
s = 0;
}
@@ -266,7 +266,7 @@ int main(int argc, char ** argv) {
std::vector<float> query_emb(n_embd, 0);
batch_decode(ctx, query_batch, query_emb.data(), 1, n_embd);
llama_batch_clear(query_batch);
common_batch_clear(query_batch);
// compute cosine similarities
{