Skip to content

allow option for using the slow tokenizer #2

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

Merged
snova-jonathanl merged 1 commit into from
Jul 12, 2024
Merged

allow option for using the slow tokenizer #2

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
3 changes: 2 additions & 1 deletion lmms_eval/models/llava_hf.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -52,6 +52,7 @@ def __init__(
device_map: str = "",
chat_template: Optional[str] = None,
use_cache: bool = True,
fast_tokenizer: bool = True,
**kwargs,
) -> None:
super().__init__()
Expand All @@ -77,7 +78,7 @@ def __init__(
self._model = LlavaForConditionalGeneration.from_pretrained(pretrained, revision=revision, torch_dtype=dtype, device_map=self.device_map, trust_remote_code=trust_remote_code, attn_implementation=attn_implementation)

self.pretrained = pretrained
self._image_processor = AutoProcessor.from_pretrained(pretrained, revision=revision, trust_remote_code=trust_remote_code)
self._image_processor = AutoProcessor.from_pretrained(pretrained, revision=revision, trust_remote_code=trust_remote_code, use_fast=fast_tokenizer)
# Pad from left for batched generation: https://huggingface.co/docs/transformers/v4.39.3/en/model_doc/llava#usage-tips
self._image_processor.tokenizer.padding_side = "left"
self._tokenizer = self._image_processor.tokenizer
Expand Down
2 changes: 1 addition & 1 deletion run_scripts/template.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -11,7 +11,7 @@ else
fi

if [[ $model_name == *"llava_hf"* ]]; then
model_args+=",device_map=auto,dtype=bfloat16"
model_args+=",device_map=auto,dtype=bfloat16,fast_tokenizer=False"
fi

echo $logger_name
Expand Down