Karol Gotkowski

Change license?

3

Hi, thanks for this project! Would it be possible to change the license to a less restrictive one? GPL is a really bad license in that sense as any project...

Temperature scaling on segmentation tasks

1

Hi, I wanted to apply temperature scaling to a segmentation task with 1 class (So a pixel belongs to this class or not). Instead of CrossEntropyLoss I am using BCEWithLogitsLoss...

Soldiers are too agressive

1

Once a soldier has found a target. There is no way to cancel his attack and order him back. No matter how many times you click to move him back,...

Mask not applied & PicklingError

4

Hi again, I tried to run the main.py but the logs do not look good... **Click to open run log** > Total Params: 1864322 Total Params: 31042434 ### SPLIT FOUND:...

This conversion script adds uncertainty / softmax / logit compression to reduce disk space consumption when saving these files. It supports three modes: 8-Bit, 16-Bit and 32-Bit compression. Even for...

Hey all, We created an integration of SAM for the python-based scientific image viewer Napari. We already extended SAM quite a bit and enable the user to: - Use SAM's...

Integration of Segment Anything into Napari

3

Hey, I have integrated Segment Anything into Napari, a scientific image viewer, and am currently implementing a 3D extension of SAM. Would you mind highlighting it in your readme? 🙂...

Hey, Related to my other issue (#8), I noticed that the MSCMR validation scribbles are missing. Could you please upload them? Best, Karol

Hey, The MSCMR dataset you provide in your repo has the wrong GT images for the train set. Instead of the actual GT images, you uploaded the scribbles under `MSCMR_dataset/tain/labels`....

Hey, I would like to make some experiments on the ACDC dataset with scribbles and would like to use the same train, val and test set as you to be...