Justus Schock

@nishant42491 sure, I'll add you to it. Open a draft PR early and ping us there if you need help :)

Hey @ankitaS11 in that case it definitely makes sense to go with your proposed way :) I updated the issue. Just let us know here, when I should assign new...

Hi @victor1cea are `ms_1` and `ms_2` of the same shape? Are `pan_1` and `pan_2` of the same shape? Do they have the same size as the ms images?

can you actually give a few sample data (just raw tensor numbers that you think should produce different outputs? I am asking because I again checked both the formula and...

@ethanwharris we had this issue already in https://github.com/PyTorchLightning/metrics/issues/74 . How does it differ or which metric does not yet allow this?

whoops, sorry I somehow overread this. I think the current design is based on the design by the losses in PyTorch (e.g. CrossEntropy), which also requires it to be integers...

> I would expect that targets in this case would be of the same dtype and shape as the preds. I don't think we should make that assumption. This would...

@benjs We discussed this as part of #1001 and likely won't enable this soon (if at all). The reason for that is two-fold: 1.) You may be able to train...

@24hours I think the way to go here would be to first try and clean up the code before we decide to dispatch to C++

@SkafteNicki I believe they are coming from resolving the merge conflicts