Roman Fitzjalen comments

Results 11 comments of


                                            Roman Fitzjalen

RuntimeError: One or more background workers are no longer alive.

Same for me. I have A100 cluster, 28 cores, 119 RAM. it works with number of precesses 4, but died if I set more. Usage of RAM for 4 processes...

Use real latest checkpoint in training

@FabianIsensee please review it. As for me I always have to rename checkpoint if I stopped it before.

MPS operator coverage tracking issue (2.6+ version)

+1 aten::max_pool3d_with_indices

nnUNetTrainer_DASegOrd0_NoMirroring error on nnunetv2==2.5.1

+ same error

Small object segmentation

Hello In my opinion patch size might play role and maybe Encoder Could you please share your training plans?

> [plans.json](https://github.com/user-attachments/files/17285257/plans.json) > > > > Here's the training plan. I also tried the EncoderM architecture, but it doesn't help much. One more important question I forgot to ask: Did...

Ensembling across different trainers for a dataset.

> Hi @aymuos15 > > > > Tricky question, you could do several things. Usually, I would also go with the most straight forward way of ensembling the folds first...

Error when using ResEncUNet with nnUNetTrainer_DASegOrd0_NoMirroring

https://github.com/MIC-DKFZ/nnUNet/issues/2480 Same issue

Tumors detected but the segmentation area is way smaller than it should be

> Hello, > > > > I have trained nnUNet on a PNET dataset. Everything works well, but there are two issues when I test the trained model on a...