Refactored Framework and Added 3D Visualization Capabilities#3
Merged
MattsonCam merged 27 commits intomainfrom Oct 29, 2025
Merged
Refactored Framework and Added 3D Visualization Capabilities#3MattsonCam merged 27 commits intomainfrom
MattsonCam merged 27 commits intomainfrom
Conversation
and changed the model to accept an input of three slices and output a one-slice segmentation
- Pass in the dataset at initialization - Ensure the tiff image suffix contains two "f"s - Seperate the cropped images from others - Moved the functionality for saving image to mlflow to a seperate util - Added an option to save images locally
- Passing in shapes instead of single dimensions.
- Added all-slice dataset for and saving utility for saving all-slices of inputs, targets, and generated predictions
- Included the ability to process the dataset that outputs all slices including:
* Selection of all-slice indices using dataset indices from the crop training dataset
* Saving images to mlflow or to a specified location
- Seperated out the functionality for retrieving whole-image metadata across all slices
versions of pytorch and cuda for the RTX 5090 GPU
|
Check out this pull request on See visual diffs & provide feedback on Jupyter Notebooks. Powered by ReviewNB |
MikeLippincott
approved these changes
Oct 16, 2025
Member
MikeLippincott
left a comment
MikeLippincott
left a comment
There was a problem hiding this comment.
LGTM. I think a little more work might be needed in documentation external of the code to explain the whole process, training approach, etc.
added 6 commits
October 16, 2025 17:17
add comments according to pr review
values based on the number of z-slices
Member
Author
|
I'm going to wait until we understand the napari/nViz visualization problem better before I merge this pr. Just in case I need to change my code |
added 4 commits
October 20, 2025 13:09
three semantic segmentation masks
to circumvent 3D rendering issues
visuzlized in napari using nviz
Member
Author
After some discussion, I think this problem has be resolved enough to merge the pr, so I am merging it now. Thanks for the review @MikeLippincott! |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This pr refactors the segmentation framework, updated the uv environment to accommodate the RTX 5090 GPU, and added 3D visualization capabilities. This pr also changes the training to accept three z-slices as input to the model, and output the middle segmentation mask. In future pr(s), I will update the framework to perform semantic segmentation and visualize those segmentations.