J. Miguel Valverde
J. Miguel Valverde
It seems that the Dice loss have been wrongly implemented. It has been defined as -dice instead of 1-dice, therefore, it will try to minimize the Dice coefficient, when it...
> 1-dice_coefficient or -dice_coeffficient makes no difference for convergence 1-dice_coef just more familiar for monitoring as its value belong to [0, 1], not [-1, 0] True, although it is a...
Hi, You can tackle class imbalance with RegionWise loss by giving less importance to the pixels that belong to a certain class (e.g., the background), and/or by giving more importance...
I don't use Windows so I cannot debug this properly, but the code and the assert condition make sense to me. It only fails on Windows right? Quite odd...
Hi, I've simply tried to access in Chrome and Firefox and I still get the same error: In both cases (Chrome and Firefox), if I access in a private tab...
Edit: Indeed, the issue appears to be what Fabian mentioned above. Quick and dirty solution: Simply by updating the file training/dataloading/nnunet_dataset.py the problem was fixed. ---- I run into this...