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: update all NextTokenChoosers

Browse Source
This commit is contained in:
drbh 2024-02-08 11:33:00 -05:00
parent 066d3d4872
commit 56e919e459
3 changed files with 3 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
server/text_generation_server/models/galactica.py
View File

@ -92,7 +92,7 @@ class GalacticaCausalLMBatch(CausalLMBatch):
requests_idx_mapping[r.id] = i
# Add escape_custom_split_sequence to the CausalLMBatch logic
inputs.append(escape_custom_split_sequence(r.inputs))
next_token_choosers.append(NextTokenChooser.from_pb(r.parameters, device))
next_token_choosers.append(NextTokenChooser.from_pb(r.parameters, device, tokenizer))
stopping_criteria = StoppingCriteria.from_pb(
r.stopping_parameters, tokenizer
)

2
server/text_generation_server/models/idefics_causal_lm.py
View File

@ -114,7 +114,7 @@ class IdeficsCausalLMBatch(Batch):
for i, r in enumerate(pb.requests):
requests_idx_mapping[r.id] = i
inputs.append(r.inputs)
next_token_choosers.append(NextTokenChooser.from_pb(r.parameters, device))
next_token_choosers.append(NextTokenChooser.from_pb(r.parameters, device, tokenizer))
stopping_criteria = StoppingCriteria.from_pb(
r.stopping_parameters, tokenizer
)

2
server/text_generation_server/models/seq2seq_lm.py
View File

@ -96,7 +96,7 @@ class Seq2SeqLMBatch(Batch):
inputs.append(r.inputs)
requests_idx_mapping[r.id] = i
decoder_input_lengths.append(1)
next_token_choosers.append(NextTokenChooser.from_pb(r.parameters, device))
next_token_choosers.append(NextTokenChooser.from_pb(r.parameters, device, tokenizer))
stopping_criteria = StoppingCriteria.from_pb(
r.stopping_parameters, tokenizer
)