Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

patch-error-on-invalid-grammar #2282

Merged
merged 3 commits into from
Jul 29, 2024
Merged

patch-error-on-invalid-grammar #2282

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
33cb2ce
quick fix
ErikKaum Jul 23, 2024
270ec41
allow silent failure
ErikKaum Jul 25, 2024
b976845
explicit todo that this is only short term
ErikKaum Jul 26, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 7 additions & 1 deletion server/text_generation_server/utils/logits_process.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -521,7 +521,13 @@ def _advance(next_token_id, fsm_grammar_state, fsm):
def _cached_compile_fsm(grammar_type, schema, tokenizer):
start_time = time.time()
if grammar_type == GrammarType.GRAMMAR_TYPE_JSON:
schema = build_regex_from_schema(schema)
try:
schema = build_regex_from_schema(schema)
# TODO: this is only here short term to avoid crashing the python server, mid term we want this in the rust/router layer
except Exception as e:
logger.error(f"Error compiling FSM, grammar won't be enforced \n{e}")
# allows everything
schema = "(.*?)"
elif grammar_type == GrammarType.GRAMMAR_TYPE_REGEX:
pass # schema is already a regex just here for clarity
fsm = RegexFSM(schema, tokenizer)
Expand Down
Loading