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

PipelineSwitchHook now compatible with MultiImageMixDataset #11716

Open
wants to merge 4 commits into
base: dev-3.x
Choose a base branch
from
Open

PipelineSwitchHook now compatible with MultiImageMixDataset #11716

Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
c856454
Add compatibility with MultiImageMixDataset
mawanda-jun May 15, 2024
081c388
fix linting
mawanda-jun May 15, 2024
3288493
fix linting
mawanda-jun May 16, 2024
dee519f
fix linting
mawanda-jun May 17, 2024
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
12 changes: 12 additions & 0 deletions mmdet/engine/hooks/pipeline_switch_hook.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -29,6 +29,18 @@ def before_train_epoch(self, runner):
# The dataset pipeline cannot be updated when persistent_workers
# is True, so we need to force the dataloader's multi-process
# restart. This is a very hacky approach.
# Make it compatible with MultiImageMixDataset
if hasattr(train_loader.dataset, 'dataset'):
# Set the loading pipeline to the identity function
# Keep it compatible with "scale_factor" key, which
# is not available otherwise.
train_loader.dataset.pipeline = Compose([lambda x: x])
# Rewrite the inner pipeline with the one given
train_loader.dataset.dataset.pipeline = \
Compose(self.switch_pipeline)
else:
train_loader.dataset.pipeline = \
Compose(self.switch_pipeline)
train_loader.dataset.pipeline = Compose(self.switch_pipeline)
if hasattr(train_loader, 'persistent_workers'
) and train_loader.persistent_workers is True:
Expand Down