Jinghong Gao comments

Results 24 comments of


                                            Jinghong Gao

about v1 how to inference in cpu?

![image](https://github.com/user-attachments/assets/8745e5ee-03ee-4d16-a312-1b219adaa964) https://github.com/MIC-DKFZ/nnUNet/tree/nnunetv1?tab=readme-ov-file#installation nnunetv1 probably doesn't support CPU.

LABEL IN DATASET.JSON WHEN I USE SPIDER DATASET

For example, if the label value is 100 as you mentioned, it should be directly converted to 8, rather than explained in the dataset.json. The same applies to 201-207; the...

Pseudo dice [0.0, 0.0, 0.0, nan, nan]

You could try reducing the learning rate.

Region-based training cannot be used with ensemble

dataset.json for coarse segmentation ```json …… "labels": { "background": 0, "myo": 1, "lv": 2, "rv": 3 }, …… ``` dataset.json for fine segmentation ```json …… "labels": { "background": 0, "myo":...

Region-based training cannot be used with ensemble

Thank you very much for your reply, it's working now.

Trying to train with GPU instead of CPU

you can try this command: CUDA_VISIBLE_DEVICES=0 nnuNetv2 train 800 3d fullres 4 --disable checkpointing -tr nnUNetTrainer 1epoch if you want to use multiple GPUs for training，you can do like this:...

Trying to train with GPU instead of CPU

Yeah, CMD can not recognize the command. You can try running command on Git bash.

Trying to train with GPU instead of CPU

try to set batch size = 1?

Trying to train with GPU instead of CPU

Can you take a complete screenshot? Start by entering commands. Try using the default trainer?

Trying to train with GPU instead of CPU

I noticed this error: "Not enough memory resources are available to process this command." How much memory do you have available when running this command? Were you able to observe...