Ross Markello

This point is well met, @Islast! Thank you so much for raising it (and thanks for pinging her in, @KirstieJane!). While `tedana` is not deterministic, we do try to ensure...

Still being considered, for sure! We're going to need to add regression tests to actually compare the outputs of the integration tests to pre-generated data, but it doesn't (didn't) make...

I do, especially if we use `nib-diff` rather than checksums. That said, I've been totally delinquent on `tedana` so I'm not sure about the status of some of the more...

You're totally right, they would both have to be in memory. I think checksums are just gonna be a real pain with floating point data, though, so I think we...

At this point I'm definitely not recommending checksums! The benefit of `nib-diff` over a straight `np.allclose` is that the former [checks things like the file header](https://github.com/nipy/nibabel/blob/master/nibabel/cmdline/diff.py), etc, to make sure...

Cool find! Just to clarify, given the number of echoes required for this do you think we should consider labelling this as "low priority?" I haven't seen any 8 echo...

@emdupre: I think @handwerkerd had updated `ctab` to just `comptable` in his branch of the code, which would also work!

Hi @Ouwen ! Most of the main image classes support a `.from_bytes()` method that can be used to load data directly into an image. (Image instances conversely also support a...

I also have a few questions that I think will help guide the rest of the PR: 1. Should users be able to include an overlay image directly in the...

Hey @effigies ! Thanks for the ping, and sorry for the delayed response. I'm happy to try and get core functionality into 3.1, but am not sure if a full...