Skip to content

fix(examples): handle missing SFT chat template #5626

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
fallintoplace wants to merge 1 commit into
base: main
Choose a base branch
from
+9 −5
Open

fix(examples): handle missing SFT chat template #5626

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
14 changes: 9 additions & 5 deletions examples/post_training/modelopt/finetune.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -135,9 +135,10 @@ def __init__(

# [WAR]: For DeepSeek-V3/R1 tokenizer, we modify the chat_template such that the <think>
# tokens are preserved for supervised learning.
self.tokenizer.chat_template = self.tokenizer.chat_template.replace(
REMOVE_THINK_CHAT_TEMPLATE, ""
)
if self.tokenizer.chat_template is not None:
self.tokenizer.chat_template = self.tokenizer.chat_template.replace(
REMOVE_THINK_CHAT_TEMPLATE, ""
)

hf_dataset_kwargs = SFTDataset.hf_dataset_to_kwargs.get(
self.hf_dataset, {"split": "train"}
Expand All @@ -158,8 +159,11 @@ def __init__(
flush=True,
)

if self.tokenizer.chat_template is None:
self.tokenizer.chat_template = SFTDataset.hf_dataset_to_prompt_template
if self.tokenizer.chat_template is None and self.hf_dataset is not None:
self.tokenizer.chat_template = SFTDataset._wildcard_get(
SFTDataset.hf_dataset_to_prompt_template,
self.hf_dataset,
)
elif self.hf_dataset is not None:
self.data_transformation = SFTDataset._wildcard_get(
SFTDataset.hf_dataset_to_conversation,
Expand Down
Loading