-
hi, I have a question. Why the size_const_cancer loss used for cancerous areas in the camelyon16 dataset, rather than size_const loss?
-
**Describe the bug**
Using overflow mode some tiles are not zero padded to the requested tile_size arg. I found this happening at /training/tumor_80.tif in CAMELYON_16 (of which the mask does not lo…
-
Hi, I noticed in the paper, you mentioned that on the camelyon16 dataset, the learning rate of classification part is multiplied by 10 until reaching 0.01, Where does it reflect in the code part?
-
Hello.
I want to test your model, but I can't find the dataset (The breast biopsy dataset) you mentioned and investigated in the paper. Can you provide a direct link to the dataset or suggest a way t…
-
**Is your feature request related to a problem? Please describe.**
Support anisotropic mpps
**Describe the solution you'd like**
read_region can be adapted to rescale the sampled region in the …
-
What is the difference among `seg_level`, `patch_level` and `custom_downsample` these three parameters?
If I want to extract features of patches in **Camelyon16** at **20x magnification**, how should…
-
Hi, I divided the patches from Cameyon16 dataset according to the CSV file in folds/ camelyon16-512-patch /split_0/fold_0/. And I also checked the input mask. But during the training process, the f1…
-
I have installed the ASAP-2.0-py38-Ubuntu2004.deb package on Ubuntu 20.04. When I try to open the .tif WSI from the Camelyon16 challenge I get the message "Invalid file type". Same happens on MacOS af…
-
First of all, thank you for the great work.
I am trying to train with your DSMIL model, but I have a different backbone that generates the embeddings, I would like to just train the MIL part. My qu…
-
Hello.
Your work is very nice and so I'm trying to reproduce your result on camelyon 16.
From WSIs (tif) of camelyon 16, I'm cropping the patches, but the number of patches (about 10000 patches …