huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-09-11 12:24:53 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix: simplify changes

Browse Source
This commit is contained in:
drbh 2024-02-28 19:54:46 +00:00
parent 4ff9cb806b
commit 0370b0feda
7 changed files with 3 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
benchmark/src/lib.rs
View File

@ -47,7 +47,6 @@ pub async fn run(
watermark, watermark,
grammar: String::new(), grammar: String::new(),
grammar_type: GrammarType::None as i32, grammar_type: GrammarType::None as i32,
grammar_state: 0,
}; };
// Initialize terminal properties // Initialize terminal properties

2
proto/generate.proto
View File

@ -80,8 +80,6 @@ message NextTokenChooserParameters {
string grammar = 10; string grammar = 10;
/// grammar type /// grammar type
GrammarType grammar_type = 11; GrammarType grammar_type = 11;
/// grammar fsm state
uint32 grammar_state = 12;
} }
message StoppingCriteriaParameters { message StoppingCriteriaParameters {

1
router/client/src/client.rs
View File

@ -130,7 +130,6 @@ impl Client {
watermark: true, watermark: true,
grammar: String::new(), grammar: String::new(),
grammar_type: GrammarType::None as i32, grammar_type: GrammarType::None as i32,
grammar_state: 0,
}), }),
stopping_parameters: Some(StoppingCriteriaParameters { stopping_parameters: Some(StoppingCriteriaParameters {
max_new_tokens: max_total_tokens - truncate, max_new_tokens: max_total_tokens - truncate,

1
router/src/health.rs
View File

@ -48,7 +48,6 @@ impl Health {
watermark: false, watermark: false,
grammar: String::new(), grammar: String::new(),
grammar_type: ProtoGrammarType::None as i32, grammar_type: ProtoGrammarType::None as i32,
grammar_state: 0,
}), }),
stopping_parameters: Some(StoppingCriteriaParameters { stopping_parameters: Some(StoppingCriteriaParameters {
max_new_tokens: 1, max_new_tokens: 1,

1
router/src/queue.rs
View File

@ -372,7 +372,6 @@ mod tests {
watermark: false, watermark: false,
grammar: String::new(), grammar: String::new(),
grammar_type: ProtoGrammarType::None as i32, grammar_type: ProtoGrammarType::None as i32,
grammar_state: 0,
}, },
stopping_parameters: StoppingCriteriaParameters { stopping_parameters: StoppingCriteriaParameters {
ignore_eos_token: false, ignore_eos_token: false,

1
router/src/validation.rs
View File

@ -356,7 +356,6 @@ impl Validation {
watermark, watermark,
grammar, grammar,
grammar_type, grammar_type,
grammar_state: 0,
}; };
let stopping_parameters = StoppingCriteriaParameters { let stopping_parameters = StoppingCriteriaParameters {
max_new_tokens, max_new_tokens,

6
server/text_generation_server/utils/tokens.py
View File

@ -468,8 +468,6 @@ class HeterogeneousNextTokenChooser:
tokenizer: PreTrainedTokenizerBase, tokenizer: PreTrainedTokenizerBase,
fsm_grammar_states: Optional[List[int]] = None, fsm_grammar_states: Optional[List[int]] = None,
) -> "HeterogeneousNextTokenChooser": ) -> "HeterogeneousNextTokenChooser":
if fsm_grammar_states is None:
fsm_grammar_states = [pb_.grammar_state for pb_ in pb]
return HeterogeneousNextTokenChooser( return HeterogeneousNextTokenChooser(
watermark=[pb_.watermark for pb_ in pb], watermark=[pb_.watermark for pb_ in pb],
temperature=[pb_.temperature for pb_ in pb], temperature=[pb_.temperature for pb_ in pb],
@ -485,7 +483,9 @@ class HeterogeneousNextTokenChooser:
tokenizer=tokenizer, tokenizer=tokenizer,
grammars=[pb_.grammar for pb_ in pb], grammars=[pb_.grammar for pb_ in pb],
grammar_types=[pb_.grammar_type for pb_ in pb], grammar_types=[pb_.grammar_type for pb_ in pb],
fsm_grammar_states=fsm_grammar_states, fsm_grammar_states=(
fsm_grammar_states if fsm_grammar_states else [0] * len(pb)
),
) )