Skip to content

Conversation

@donglaiw
Copy link
Collaborator

  • Add do_2d parameter to DataConfig and InferenceDataConfig
  • Modify MonaiConnectomicsDataset to handle 2D dimensions when do_2d=True
  • Update model wrapper to squeeze/unsqueeze depth dimension for 2D models
  • Disable sliding window inference for 2D models, use direct inference
  • Make TTA flip axes dynamic based on data dimensions (2D vs 3D)
  • Switch evaluation metric from Jaccard to Adapted Rand Error for instance segmentation
  • Update monai2d_worm.yaml config to enable 2D processing

This enables seamless 2D data processing while maintaining 3D compatibility.

- Add do_2d parameter to DataConfig and InferenceDataConfig
- Modify MonaiConnectomicsDataset to handle 2D dimensions when do_2d=True
- Update model wrapper to squeeze/unsqueeze depth dimension for 2D models
- Disable sliding window inference for 2D models, use direct inference
- Make TTA flip axes dynamic based on data dimensions (2D vs 3D)
- Switch evaluation metric from Jaccard to Adapted Rand Error for instance segmentation
- Update monai2d_worm.yaml config to enable 2D processing

This enables seamless 2D data processing while maintaining 3D compatibility.
@donglaiw donglaiw merged commit 5a719b4 into PytorchConnectomics:master Oct 29, 2025
0 of 14 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant