Skip to content

support qwen2.5-vl in vllm_grpo_trainer #104

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

Open
zhangzef wants to merge 1 commit into
base: main
Choose a base branch
from
Open

support qwen2.5-vl in vllm_grpo_trainer #104

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
17 changes: 13 additions & 4 deletions src/r1-v/src/open_r1/trainer/vllm_grpo_trainer.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -36,6 +36,7 @@
PreTrainedModel,
PreTrainedTokenizerBase,
Qwen2VLForConditionalGeneration,
Qwen2_5_VLForConditionalGeneration,
Trainer,
TrainerCallback,
is_wandb_available,
Expand Down Expand Up @@ -175,6 +176,10 @@ def __init__(
model = Qwen2VLForConditionalGeneration.from_pretrained(
model, **model_init_kwargs
)
elif "Qwen2.5-VL" in model_id:
model = Qwen2_5_VLForConditionalGeneration.from_pretrained(
model, **model_init_kwargs
)
elif "Aria" in model_id:
model_init_kwargs.pop("use_cache")
model = AriaForConditionalGeneration.from_pretrained(
Expand All @@ -199,6 +204,10 @@ def __init__(
self.ref_model = Qwen2VLForConditionalGeneration.from_pretrained(
model_id, **model_init_kwargs
)
elif "Qwen2.5-VL" in model_id:
self.ref_model = Qwen2_5_VLForConditionalGeneration.from_pretrained(
model_id, **model_init_kwargs
)
elif "Aria" in model_id:
self.ref_model = AriaForConditionalGeneration.from_pretrained(
model_id, **model_init_kwargs
Expand All @@ -217,12 +226,12 @@ def __init__(

# Processing class
if processing_class is None:
if "Qwen2-VL" in model_id or "Aria" in model_id:
if "Qwen2-VL" in model_id or "Aria" in model_id or "Qwen2.5-VL" in model_id:
processing_class = AutoProcessor.from_pretrained(model_id)
pad_token_id = processing_class.tokenizer.pad_token_id
processing_class.pad_token_id = pad_token_id
processing_class.eos_token_id = processing_class.tokenizer.eos_token_id
if "Qwen2-VL" in model_id:
if "Qwen2-VL" in model_id or "Qwen2.5-VL" in model_id:
processing_class.image_processor.max_pixels = max_pixels
processing_class.image_processor.min_pixels = min_pixels
else:
Expand Down Expand Up @@ -304,12 +313,12 @@ def data_collator(features): # No data collation is needed in GRPO
# rewrite the processing AutoTokenizer -> AutoProcessor
model_id = model if isinstance(model, str) else model.config._name_or_path
if processing_class is None:
if "Qwen2-VL" in model_id or "Aria" in model_id:
if "Qwen2-VL" in model_id or "Aria" in model_id or "Qwen2.5-VL" in model_id:
processing_class = AutoProcessor.from_pretrained(model_id)
pad_token_id = processing_class.tokenizer.pad_token_id
processing_class.pad_token_id = pad_token_id
processing_class.eos_token_id = processing_class.tokenizer.eos_token_id
if "Qwen2-VL" in model_id:
if "Qwen2-VL" in model_id or "Qwen2.5-VL" in model_id:
processing_class.image_processor.max_pixels = max_pixels
processing_class.image_processor.min_pixels = min_pixels
else:
Expand Down