Fixing errors in CI

[snbianco]

Updating the notebooks that are failing in our nightly CI build.

[snbianco]

Tagging @adrianlucy @scfleming for visibility

[adrianlucy]

Thanks Sam! Will review ~Thursday-ish, but one quick minor thing that caught my eye: diff changes from "execution_count": null to e.g. "execution_count": 1 suggest some cells may be executed in this commit? I think preference is for notebooks to be unexecuted in this repo, mostly (I think?) to save on space, also makes git diffs easier to read, maybe other reasons that no one ever told me.

In theory if the unwanted output in this PR takes up significant storage, I guess you might want to consider squashing the commits that take up space in the history?, but up to you / judgement call whether that's worth your time.

[snbianco]

Thanks Sam! Will review ~Thursday-ish, but one quick minor thing that caught my eye: diff changes from "execution_count": null to e.g. "execution_count": 1 suggest some cells may be executed in this commit? I think preference is for notebooks to be unexecuted in this repo, mostly (I think?) to save on space, also makes git diffs easier to read, maybe other reasons that no one ever told me.

In theory if the unwanted output in this PR takes up significant storage, I guess you might want to consider squashing the commits that take up space in the history?, but up to you / judgement call whether that's worth your time.

That is weird, apparently the "Clear All Outputs" button in VSCode doesn't actually reset the execution count! The notebooks shouldn't actually have any output. I'll see if I can find a way to reset that variable.

[adrianlucy]

Oh interesting, weird! At least we don't have to worry about the commit history then. FWIW in Jupyter itself I think it's "restart kernel and clear outputs"(?) but I can see it being a hassle to switch between VScode's implementation and Jupyter.