forked from AtWillB/Unet_Spokes
-
Notifications
You must be signed in to change notification settings - Fork 0
Yugadeep/spokes_ML
Folders and files
Name | Name | Last commit message | Last commit date | |
---|---|---|---|---|
Repository files navigation
#YK: The repo in its current state is intended for my research groupmates. But, in the future, I hope to make it more accessible and explain all the steps more clearly. (Dec 26th, 2024) Steps I'm taking to make an ML model to detect spokes on images: # obviously, I learned most of these steps from Will's work 1) Choose a training set of rpjbs (images from Dr. Hedman's group), process them into thumbanils of (736, 160) size. 2) Label them i.e., draw masks on the thumbnails on label-studio. 2) Export masks from label-studio as PNGs, but their names are ruined in the process. So, export masks again as a json file. Rename PNGs using the json file. 3)*** MAKE SURE THAT THE MASKS AND THUMBNAILS ARE THE SAME SIZE ***. The unet framework expects the same size between masks and thumbnails. And also, that they shape is divisible by a specific number. 4) Run model_generator.ipynb notebook if you want to make a model in the notebook. Or run mkmodel.py if you want to run it on the terminal.
About
Copying Will's personal spokes repo to make edits and changes.
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published
Languages
- Jupyter Notebook 97.3%
- Python 2.7%