vmiller987

I had some time this morning to work through this issue a bit. I think the issue might have to do with nnUNet being compiled on a FIPS disabled machine....

Hello @FabianIsensee @mrokuss, Would you happen to know a direction or where to start for an issue like this? We would like to be able to use nnUNet as intended....

I haven't tested @rubencardenes code yet, but I wanted to add I would also be interested in an ONNX export, as I'm unable to install and use nnUNet due to...

@rubencardenes > Hi, conversion to ONNX can be done like this: > > def convert_to_ONNX( > > Hope it helps Hey, I was able to make this work and get...

@cyber-chris Do you mind sharing your solution? Unfortunately, if you make a PR it won't be looked at. They are very busy with other projects.

@cyber-chris Thanks! I will give this a try later this week. I have images that are (512, ~2000, 512), with spacings of (0.6, 0.9, 0.6) and I've been thinking there's...

They have fixed this issue on master. Try `pip install git+https://github.com/MIC-DKFZ/nnUNet.git` and see if this works. It's resolved my preprocessing errors. You were certainly correct about blosc2.

Using torch 2.8.0 and 2.9.1, and trying to run the example on the home page results in the same error: ```python import io import requests import supervision as sv from...

> Should work in most recent branch, don't think the fix has made it into a release yet I have confirmed master works. ```bash uv venv --python 3.13 source .venv/bin/activate...

For training, `nnUNet_compile=False` helps, but I also have to do `nnUNet_n_proc_DA=0` every single time or I get the background workers die when training. I cannot run it without this line.