Skip to content
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

Jump to bottom

[bugfix] remove nan loss in manual optimization #5121

Merged
merged 7 commits into from
Dec 16, 2020
Merged

[bugfix] remove nan loss in manual optimization #5121

Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
3bca9a0
remove nan loss whe missing
tchaton Dec 14, 2020
4e09e31
Merge branch 'master' into bugfix/manual_optimizaiton_loss_nan
tchaton Dec 14, 2020
a98d3f7
Update pytorch_lightning/core/lightning.py
tchaton Dec 15, 2020
af4fb09
Merge branch 'master' into bugfix/manual_optimizaiton_loss_nan
tchaton Dec 15, 2020
3688709
Apply suggestions from code review
rohitgr7 Dec 15, 2020
2d46843
Merge branch 'master' into bugfix/manual_optimizaiton_loss_nan
rohitgr7 Dec 15, 2020
5bb7406
Merge branch 'master' into bugfix/manual_optimizaiton_loss_nan
tchaton Dec 16, 2020
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
15 changes: 9 additions & 6 deletions pytorch_lightning/core/lightning.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1392,12 +1392,15 @@ def get_progress_bar_dict(self):
"""
# call .item() only once but store elements without graphs
running_train_loss = self.trainer.train_loop.running_loss.mean()
avg_training_loss = (
running_train_loss.cpu().item()
if running_train_loss is not None
else float("NaN")
)
tqdm_dict = {"loss": "{:.3g}".format(avg_training_loss)}
avg_training_loss = None
if running_train_loss is not None:
avg_training_loss = running_train_loss.cpu().item()
elif self.trainer.train_loop.automatic_optimization:
avg_training_loss = float('NaN')

tqdm_dict = {}
if avg_training_loss is not None:
tqdm_dict["loss"] = f"{avg_training_loss:.3g}"

if self.trainer.truncated_bptt_steps is not None:
tqdm_dict["split_idx"] = self.trainer.split_idx
Expand Down