Kenneth C. Arnold

See https://www.together.xyz/blog/neurips-2022-overcoming-communication-bottlenecks-for-decentralized-training-12 for a discussion of how to make this practical.

Thoughtful comment! [This paper](https://arxiv.org/abs/2103.10427) argues that a low-rank bias is actually helpful, and that even back-to-back supposedly-extraneous linear layers end up implicitly regularizing networks towards that state.

This looks like the same issue as https://github.com/fastai/fastai/issues/3861.

For whatever it's worth, I just got an Apple silicon mac and the Homebrew quodlibet doesn't work, but following the [dev instructions](https://quodlibet.readthedocs.io/en/latest/development/devenv.html#devenv) ("If you want to run the tests with...

Since the original author said they won't maintain this anymore, could you publish this fixed version? Until then, others can get this working by downloading [this ZIP file](https://codeload.github.com/Iksas/media-import/zip/refs/heads/pyqt6-fix) and dragging...

Loading the model succeeds, but running any eval code fails because there's `.cuda()` calls all over the place. You'll have to remove all of them. I made a quick fork...

I encountered this problem with nbformat 5.4.0 installed. Upgrading nbformat (to 5.7.3) made the error message much more helpful. I think the "confusing error message" part of this bug is...

I've also seen various other duplicated blocks, *not* on renamed pages. So perhaps it's actually unrelated to renaming, or perhaps renaming makes triggering the bug more likely. Unfortunately this means...

This issue would apply to anything that uses `htmlwidgets` in `learnr` then, not just maps, right? If so, then manual patches like this will be needed for other types of...

I was also surprised by newlines getting stripped. I expected this code to modify just the frontmatter, not the rest of the file.