Open xpetitt123 opened 3 months ago
I should try to write a shell script in order to dump every image in a single .tar
file and then use the CNN.
The shell script is hopefully done and working and now we can proceed with the analysis of the entire dataset at once.
Continue with the image analysis pipeline, explore more the CNN model that was used, implement different dimensionality reduction algorithms and clustering methods. Also explore the possibility of persistent images or directly extracting persistent homology values from the pixel resolutions of the images instead.
We could also try to play around with Mapper for the clustering.