Carsten L

Hi @farrell236, thank you for looking into this. Since the images we are working with are large in spatial resolution the RAM is always an issue so your proposals are...

Hi @Hollycooool, to write out the softmax probabilities the network returns alongside the segmentations during the prediction use the `--save_probabilties` flag. This will save `.npz` files in the ouptut folder...

Hi @Hollycooool To get the logits of nnU-Net you will have to change the inference code of nnU-Net. As of yet we do not support this. --save_probabilities saves the probabilities...

Hi @Hollycooool, The necessary steps are detailed in the [README.md](https://github.com/MIC-DKFZ/nnUNet/tree/master/nnunetv2/inference) in the folder `nnunetv2/inference `. Best regards, Carsten

Hi @clarkbab, I am not sure whether I correctly understand you but nnU-Net has as any other training based models issue if structures which are supposed to be annotated are...

Hi @clarkbab, The documentation states that not all labels have to be present inside of each image. Given the following 3 classes: ``` 0: background 1: brain 2: brain tumor...

Hi clarkbab, the output of nnU-Net before postprocessing is a 4D logit map. So an approach like this is possible, and something similar has been implemented for the following paper:...

Hi @NathanMolinier, thanks for getting in touch with us regarding your work on this. It looks very promising! As your code is nicely integrated into nnU-Net people can already choose...

Hi @jhdezr1, The simplest way to do so is by adding a custom plans file and using these plans for training. You can do so by writing a custom planner...

Hi @Boboshinidie, the official documentation for preprocessing guidelines for nnU-Net can be found here: https://github.com/MIC-DKFZ/nnUNet/blob/master/documentation/dataset_format.md Further, we provide an example for the 2d use-case here: https://github.com/MIC-DKFZ/nnUNet/blob/master/nnunetv2/dataset_conversion/Dataset120_RoadSegmentation.py Best regards, Carsten