Fernando Pérez-García

Awesome 💯 Happy to help if needed. I'll add the intensity kwarg soon.

> @fepegar would it be easy to add Motion transform in the Slicer pluggin ? (this would answer the initial need of exploration) and more generally it may be interesting...

Hi, @mathpluscode. Nice to see you here. Would the example in https://github.com/fepegar/torchio/issues/205#issuecomment-647703444 work? If the reading time is another bottleneck, we could implement a reader that uses NiBabel's `memmap` to...

Nice usage of samplers! Thanks, @mathpluscode.

@jxchen01 it would be nice to have this, yes. That's why I never closed the issue. Using a `WeightedSampler` (the parent of `LabelSampler`) might make training slow, but using a...

Sure, I guess that could work too, and might be faster. But to be consistent with the rest of the library, it should be a new transform called `RandomCrop` that...

I was rather thinking of leaving `CropOrPad` as is and creating a `RandomCrop` that computes the needed `cropping` argument for `CropOrPad`.

Hey, @jxchen01. I've seen your PR. When would you need random padding? Why not simply implementing random cropping as in https://github.com/fepegar/torchio/issues/570#issuecomment-860265126? Moreover, why not using patch-based training instead?

Hi, @justusschock. Thanks for the suggestion. > What file formats do we want to support? I think for images we definitely need multiple (especially nifti and dicom) but for metadata...

If you both think it would be useful, I'm happy to review your PR!