Skip to content

feat: Ensure that training resumes from latest checkpoint #63

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
linnanwang merged 5 commits into from
Nov 20, 2025
Merged

feat: Ensure that training resumes from latest checkpoint #63

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
530357c
feat: Ensure that training resumes from latest checkpoint
pthombre Nov 19, 2025
5b6fdf7
Fix lint issues
pthombre Nov 19, 2025
50343c5
Fix import order
pthombre Nov 19, 2025
e2c7ab8
Merge branch 'main' into pranav/resume_training_automodel
linnanwang Nov 19, 2025
fdd3af4
Merge branch 'main' into pranav/resume_training_automodel
abhinavg4 Nov 20, 2025
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
5 changes: 2 additions & 3 deletions dfm/src/automodel/recipes/train.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -306,9 +306,8 @@ def setup(self):
start_epoch=int(self.start_epoch),
num_epochs=int(self.num_epochs),
)
# Optional resume only through config-defined restore_from
if self.restore_from:
self.load_checkpoint(restore_from=self.restore_from)

self.load_checkpoint(self.restore_from)

if is_main_process():
os.makedirs(self.checkpoint_config.checkpoint_dir, exist_ok=True)
Expand Down
2 changes: 1 addition & 1 deletion examples/automodel/finetune/finetune.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -19,7 +19,7 @@
from dfm.src.automodel.recipes.train import TrainWan21DiffusionRecipe


def main(default_config_path="/opt/DFM/dfm/examples/Automodel/finetune/wan2_1_t2v_flow.yaml"):
def main(default_config_path="examples/automodel/finetune/wan2_1_t2v_flow.yaml"):
cfg = parse_args_and_load_config(default_config_path)
recipe = TrainWan21DiffusionRecipe(cfg)
recipe.setup()
Expand Down
2 changes: 1 addition & 1 deletion examples/automodel/pretrain/pretrain.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -19,7 +19,7 @@
from dfm.src.automodel.recipes.train import TrainWan21DiffusionRecipe


def main(default_config_path="/opt/DFM/dfm/examples/Automodel/pretrain/wan2_1_t2v_flow.yaml"):
def main(default_config_path="examples/automodel/pretrain/wan2_1_t2v_flow.yaml"):
cfg = parse_args_and_load_config(default_config_path)
recipe = TrainWan21DiffusionRecipe(cfg)
recipe.setup()
Expand Down
Loading