Multichannel Timeseries

[droumis]

[github-actions]

Your changes were successfully integrated in the dev site, make sure to review
the pages of the projects you touched before merging this PR: https://holoviz-dev.github.io/examples/.
You can also download an archive of the site from the workflow summary page which comes in handy
when your dev site built was overriden by another PR (we have a single dev site!).

[github-actions]

Your changes were successfully integrated in the dev site, make sure to review
the pages of the projects you touched before merging this PR: https://holoviz-dev.github.io/examples/.
You can also download an archive of the site from the workflow summary page which comes in handy
when your dev site built was overriden by another PR (we have a single dev site!).

[droumis]

Just noting that for the multichan.ipynb, I'm having to use an Overlay instead of e.g. curves_overlay = hv.NdOverlay(curves, 'Channel', sort=False)... I wonder if this impacts performance at all

[droumis]

Note: Check to ensure that the resultant downsampled data pyramid doesn't ever live entirely in memory prior to getting written to disk.

[droumis]

Note: find better principled way to determine factors for creating the data pyramids..
The following doesn't work as the number of channels scales
FACTORS = list(np.array([1, 2, 4, 8, 16, 32, 64, 128, 256]) ** (len(ts_ds["channel"]) // 4))

[github-actions]

Your changes were successfully integrated in the dev site, make sure to review
the pages of the projects you touched before merging this PR: https://holoviz-dev.github.io/examples/.
You can also download an archive of the site from the workflow summary page which comes in handy
when your dev site built was overriden by another PR (we have a single dev site!).

[github-actions]

Your changes were successfully integrated in the dev site, make sure to review the pages of the projects you touched before merging this PR.

[github-actions]

Your changes were successfully integrated in the dev site, make sure to review the pages of the projects you touched before merging this PR.

[maximlt]

I think the bottom of this cell is a good place to add f.close().

[droumis]

f is only used in conversion to xarray. I added a context manager instead of f.close()

[maximlt]

Arf now I get this error when the pyramid gets created, while I don't when close is called afterwards. I'll make the change.

OSError: Can't synchronously read data (dset_id is not a dataset ID)

[maximlt]

There are still some unresolved comments, specially for the large_multichan notebook.

[droumis]

@maximlt, I think I've addressed everything identified up until now. Sorry for not seeing some of the previous review comments until recently. Thanks for the epic review!

[github-actions]

Your changes were successfully integrated in the dev site, make sure to review the pages of the projects you touched before merging this PR.

[github-actions]

Your changes were successfully integrated in the dev site, make sure to review the pages of the projects you touched before merging this PR.

[maximlt]

Thanks for the epic review!

We got well over-trained to do this last year with the NF SDF grant! Thanks for adjusting the example based on our feedback, it's such a cool one. It should totally be in the Featured section, except that I think keeping 8 examples there sounds reasonable and I didn't which one to remove, so that's for later :) Merging, congrats!

[jbednar]

I agree that 8 is the max for that section. It's a very tough call; those are all very cool examples, but if I think about what they demonstrate, this example covers things not included there at all, so I do think it should bump something out. I'd guess we should replace Exoplanets in that group, given that it's not currently showing any plots (which is maybe an error?).