CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-10-31 08:51:55 +00:00
Code Issues Packages Projects Releases Wiki Activity

examples : add GBNF validator program (#5948)

Browse Source 
* Revising GBNF validator program to be much simpler.

* Changing from streams to using cstdio

* Adding final newline character.
This commit is contained in:
Clint Herron
2024-04-04 03:44:28 -04:00
committed by GitHub
parent 4399f13fb9
commit 9b84ae1806
5 changed files with 171 additions and 20 deletions
Download Patch File Download Diff File Expand all files Collapse all files

22
llama.cpp
View File

@@ -11621,28 +11621,10 @@ static std::vector<llama_vocab::id> llama_tokenize_internal(const llama_vocab &
// grammar - internal
//
struct llama_partial_utf8 {
uint32_t value; // bit value so far (unshifted)
int n_remain; // num bytes remaining; -1 indicates invalid sequence
};
struct llama_grammar {
const std::vector<std::vector<llama_grammar_element>> rules;
std::vector<std::vector<const llama_grammar_element *>> stacks;
// buffer for partially generated UTF-8 sequence from accepted tokens
llama_partial_utf8 partial_utf8;
};
struct llama_grammar_candidate {
size_t index;
const uint32_t * code_points;
llama_partial_utf8 partial_utf8;
};
// Decodes a UTF-8 string which may end in an incomplete sequence. Adds a terminating 0 for use as
// pointer. If an invalid sequence is encountered, returns `llama_partial_utf8.n_remain == -1`.
static std::pair<std::vector<uint32_t>, llama_partial_utf8> decode_utf8(
std::pair<std::vector<uint32_t>, llama_partial_utf8> decode_utf8(
const std::string & src,
llama_partial_utf8 partial_start) {
static const int lookup[] = { 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 2, 2, 3, 4 };
@@ -11844,7 +11826,7 @@ static void llama_grammar_advance_stack(
// be positioned at a character range (see `llama_grammar_advance_stack`), and
// produces the N possible stacks if the given char is accepted at those
// positions
static std::vector<std::vector<const llama_grammar_element *>> llama_grammar_accept(
std::vector<std::vector<const llama_grammar_element *>> llama_grammar_accept(
const std::vector<std::vector<llama_grammar_element>> & rules,
const std::vector<std::vector<const llama_grammar_element *>> & stacks,
const uint32_t chr) {