Skip to content
This repository has been archived by the owner on Mar 21, 2024. It is now read-only.

Fix multi-node bug in PL 1.2.8 #437

Merged
merged 9 commits into from
Apr 16, 2021
Merged

Fix multi-node bug in PL 1.2.8 #437

Changes from 1 commit
Commits
Show all changes
9 commits
Select commit Hold shift + click to select a range
ca8ea84
Fix the bug in PL
Apr 16, 2021
f2ea55c
Add back the test
Apr 16, 2021
3374b97
Missing import
Apr 16, 2021
c37a423
CHANGELOG.md
Apr 16, 2021
93b0cee
Fix it
Apr 16, 2021
fa179a0
Only plugin if more than one gpu
Apr 16, 2021
9751d79
Only plugin if more than one gpu
Apr 16, 2021
68db907
Mypy
Apr 16, 2021
940cb1c
Mypy again
Apr 16, 2021
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
Prev Previous commit
Next Next commit
Fix it
  • Loading branch information
melanibe committed Apr 16, 2021
commit 93b0ceef726b321eea9d59cadf7fa7a7a3a3ad6f
2 changes: 1 addition & 1 deletion InnerEye/ML/model_training.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -145,7 +145,7 @@ def create_lightning_trainer(config: ModelConfigBase,
sync_batchnorm=True,
terminate_on_nan=config.detect_anomaly,
resume_from_checkpoint=str(resume_from_checkpoint) if resume_from_checkpoint else None,
plugins=[InnerEyeDDPPlugin(num_nodes=2, sync_batchnorm=True)]
plugins=[InnerEyeDDPPlugin(num_nodes=num_nodes, sync_batchnorm=True)]
)
return trainer, storing_logger

Expand Down