Dan Handwerker

I think/hope this is fully done and ready for review. It's important to have people look at the code, but I'd also like to have people read the documentation (`tedana...

> I really don't want to hold this PR up any more. I say we merge as-is, make a release-candidate release, and deal with any bugs or typos before the...

I'm at a conference 8/6-9 so 8/12 would be a good day to merge if we don't get feedback requiring changes. (Earlier if we get feedback saying everything is fine....

Thank you all. I'm comfortable merging, but I'm currently traveling through Tuesday and wouldn't mind giving it one more quick check myself. If others are comfortable merging, then give approving...

From our April 2024 dev call, the decision is that the function has fine behavior, but we should update the document to accurately reflect what the function can do.

The code looks fine to me. I don't love that `1.0` (float) will output 100% variance and `1` (int) will output 1 component, but the documentation is clear and anyone...

Every time I look at the code, I think of ways to do things better, but I think this PR is fine. What's happening is logged fairly clearing in the...

I don't think your idea for `--tedpca 0` or `--tedpca all` would work. If there are 200 volumes and you send 200 components to the ICA step, it would never...

I'm a bit confused on wanting `n_components == min(n_samples, n_features) - 1` Whenever I've ended up with that ICA has serious trouble converging since there are too many components. Even...