Bin Li
Bin Li
In `simclr/run.py`: https://github.com/binli123/dsmil-wsi/blob/e637f8295e0bc580e20569586b11ce69a75190c6/simclr/run.py#L8-L19 Basically, these are paths to all the patches.
Which script did you use for generating the prediction outputs? It could be that the trained model weights are not loaded correctly.
#44 Maybe these two threads should be put together?
Ideally no. You could try to exclude the patches extracted from the testing images, but the impact is probably minor since the number of patches from the training set alone...
https://drive.google.com/drive/folders/1_mumfTU3GJRtjfcJK_M0fWm048sYYFqi There are several model weights trained using only the training data. I also tested using both the training set and the testing set for SimCLR, the difference in the...
In my case both are converging, v2 converges significantly slower at around 100 epoch. If you load "init.pth" it would be much faster.
Please check out the folder [`simclr`](https://github.com/binli123/dsmil-wsi/tree/master/simclr)
> Thanks for sharing this amazing work. I wonder what is the 'init.pth' for the dsmil model. Is it trained on some specific dataset? > > Do I need to...
> Hi, > > Thanks for sharing your code. I wanted to concatenate two sets of features that I extracted from my dataset (both at the same magnification) using compute_feats.py,...
You need only to swap in the ABMIL aggregation head which takes precomputed feature vectors as inputs.