OASIS Brain MRI Stable Diffusion

[jonooallen]

Requesting to submit my Stable Diffusion model into PatternFlow. This model is a stable diffusion model trained on the OASIS Brain MRI data. Overall this model performed excellently, and can now generate Brain MRI from noise. It uses a cross attention positional embedded U-Net to add and remove noise in training. This learned removal of noise can then be run on random noise images to generate new Brain MRIs. Further details about the project, model, and results can be found in my readme.

[SiyuLiu0329]

This is an initial inspection, no action is required at this point

• Readme: OK but some result visualisation would be good. A full dependency dump is usually not the most helpful as most packages are trivial. Please list key dependencies in the future.
• Results: OK, model is working. Please insert generated images into the readme directly, external links can expire and can be hard to see.
• Code: OK, but i think you trained it on the segmentation maps instead of the images.
• Commit messages: OK but some should be more specific

[shakes76]

Good Practice (Design/Commenting, TF/Torch Usage)

Adequate use and implementation
Good spacing and comments
Header blocks missing -1

Recognition Problem

Solves problem (used label maps instead of images) -1
Driver Script present
File structure present
Shows Usage & Demo & Visualisation & Data usage
Module present
Commenting
No Data leakage
Difficulty: Hard

Commit Log

Meaningful commit messages (could be more informative) -1
Progressive commits used

Documentation

ReadMe acceptable/good
Good Description and Comments
Markdown used PDF submitted

Pull Request

Successful Pull Request (Working Algorithm Delivered on Time in Correct Branch)
Feedback required, revert changes to other student files -2
Request Description OK, could be more descriptive of your contribution -1