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

Fix bug comparing max_steps to global step which inits at 0 #4278

Merged
merged 10 commits into from
Oct 22, 2020
Merged

Fix bug comparing max_steps to global step which inits at 0 #4278

Changes from 1 commit
Commits
Show all changes
10 commits
Select commit Hold shift + click to select a range
9b47abd
Fix bug comparing max_steps to global step which inits at 0
SeanNaren Oct 21, 2020
1697b0a
Added test to ensure accumulate grad batch works with max steps
SeanNaren Oct 21, 2020
720472b
Merge branch 'master' into bug/4193-accum
awaelchli Oct 22, 2020
fd7724f
check fix with TODO test
awaelchli Oct 22, 2020
85d9e29
correct call counts
awaelchli Oct 22, 2020
18fb249
Add check to ensure we've finished accumulation of this global step b…
SeanNaren Oct 22, 2020
27612ff
Remove + 1 check in test as this was incorrect
SeanNaren Oct 22, 2020
122ab38
Update incorrect expected outputs in lr finder test
SeanNaren Oct 22, 2020
4b73d44
Added brackets for clarity
SeanNaren Oct 22, 2020
823ccc9
Merge branch 'master' into bug/4193-accum
SeanNaren Oct 22, 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
2 changes: 1 addition & 1 deletion pytorch_lightning/trainer/trainer.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -481,7 +481,7 @@ def train(self):
# run train epoch
self.train_loop.run_training_epoch()

if self.max_steps and self.max_steps <= self.global_step:
if self.max_steps and self.max_steps <= self.global_step + 1:
SeanNaren marked this conversation as resolved.
Show resolved Hide resolved

# hook
self.train_loop.on_train_end()
Expand Down