Skip to content

[bugfix] Fix args template type #8005

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
Jintao-Huang merged 2 commits into from
Feb 7, 2026
Merged

[bugfix] Fix args template type #8005

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
3bf1784
support deepseek v3_2 megatron
ccyhxg Feb 6, 2026
c2ed79c
fix template_type
ccyhxg Feb 7, 2026
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
2 changes: 1 addition & 1 deletion requirements/framework.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -34,7 +34,7 @@ sortedcontainers>=1.5.9
tensorboard
tiktoken
tqdm
transformers>=4.33,<4.58
transformers>=4.33
Copy link
Contributor

@gemini-code-assist gemini-code-assist bot Feb 7, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

Removing the upper version constraint for transformers can introduce instability if future releases include breaking changes. It's a best practice to pin dependencies with an upper bound to ensure reproducibility and prevent unexpected failures. Please consider adding an upper bound, for instance, <5.0 or based on the latest version you've tested against.

transformers>=4.33,<5.0

transformers_stream_generator
trl>=0.15,<0.25
uvicorn
Expand Down
3 changes: 1 addition & 2 deletions swift/callbacks/perf_log.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5,7 +5,7 @@
import torch
from transformers import TrainerControl, TrainerState

from swift.utils import empty_cache, get_logger
from swift.utils import empty_cache, get_current_device, get_device_count, get_env_args, get_logger
from .base import TrainerCallback

if TYPE_CHECKING:
Expand Down Expand Up @@ -43,7 +43,6 @@ def __init__(self, args: 'TrainingArguments', trainer: 'Trainer'):
self.step_start_time = None

def on_init_end(self, args: 'TrainingArguments', state: TrainerState, control: TrainerControl, **kwargs):
from swift.utils import get_current_device, get_device_count, get_env_args

# Top priority. Specify by ENV
tflops = get_env_args('DEVICE_TFLOPS', int, None)
Expand Down
2 changes: 1 addition & 1 deletion swift/template/register.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -25,7 +25,7 @@ def _read_args_json_template_type(model_dir):
return
from swift.arguments import BaseArguments
args = BaseArguments.from_pretrained(model_dir)
return args.template_type
return args.template


def get_template_meta(model_info: 'ModelInfo',
Expand Down
Loading