Expanding the `hvsampledata` Repository with More Datasets

[Azaya89]

As part of the NumFocus SDG project focused on revitalizing the hvPlot documentation website, I have been working on identifying all datasets currently used in the hvPlot documentation. I have also compiled a new spreadsheet suggesting datasets that could be included in the hvsampledata repository, which the new hvPlot documentation will reference.

However, this repository is not intended for hvPlot alone; it is meant to serve broader Holoviz projects, as discussed in holoviz/holoviz#394. To ensure it effectively meets the needs of various projects within the Holoviz ecosystem, I would appreciate additional suggestions and contributions for datasets that could be added.

You can view the current list of datasets and proposed additions in this spreadsheet:
➡️ hvsampledata spreadsheet

How to Contribute

• Review the datasets currently listed in the spreadsheet.
• Suggest new datasets that align with the repository's goals.
• Ensure suggestions follow the existing dataset format for consistency.
• Leave comments on the spreadsheet or respond to this issue with your recommendations.

Thank you for your contributions!

@holoviz/holoviz-dev

[maximlt]

Having a points location dataset would be useful, with these properties ideally:

• lightweight
• CSV (easy to read)
• already used by a few HoloViz projects
• lat/lon
• a couple of categorical columns
• a couple of numeric columns (for e.g. bubble map)

EDIT:

• spatial extent: I'd say wide enough so we can show how the data looks when projected

cc @Azaya89

[Azaya89]

I added the world cities dataset to the spreadsheet. It:

• is a csv file
• has lat/lon columns
• can be made lightweight
• has some categorical columns (cities and countries)
• has a numerical column (population)
• spatial enough (covers cities around the world)

This dataset can replace the airports dataset, especially as that one does not have a good numerical column that can be useful for bubble charts. It also covers more worldwide locations that the airports dataset, which is mostly focused on north america.