lsst-epo / citizen-science-notebooks

A collection Jupyter notebooks that can be used to associate Rubin Science Platform data to a Zooniverse citizen science project.
3 stars 1 forks source link

Move things to zooniverse #21

Open bnord opened 1 year ago

bnord commented 1 year ago

tabular, flipbook; lightcurve, associated images?, send lightcurve/image

ericdrosas87 commented 1 year ago

Sending tabular data is currently possible with the Zooniverse API and a little Python added to the notebooks, as is sending a lightcurve image.

I think it's the flipbook and lightcurve-sent-as-tabular-data-but-rendered-as-graph functionality that we're waiting on Zooniverse to make available on their platform by default.

eatyourgreens commented 1 year ago

Regarding data rendered as a graph, we support subjects that use a .json URL, with the application/json MIME type, as long as the content of the JSON payload conforms to a format that we support.

I've just published an Architecture Decision Record for JSON subjects in the new Zooniverse front-end: https://github.com/zooniverse/front-end-monorepo/blob/master/docs/arch/adr-48.md

I'm happy to explain what we expect to see in the payload too. For an example of a project with JSON subjects, have a look at the subjects for SuperWASP: Black Hole Hunters. For example: https://panoptes-uploads.zooniverse.org/subject_location/f9c8bdd4-f545-4c3f-b03e-f4fd7ba78000.json

I've also got examples of x,y data with multiple series and error bars, for supernova candidates, that I can share.

eatyourgreens commented 1 year ago

If it's helpful, I've also got some Python code that I've been using to generate JSON subjects from ZTF data. LSST data should work if it follows the same schema: https://github.com/zooniverse/Zooniverse_SLSN/blob/69920904a6723d53fe2e04c4248353acaa84f79f/lasair_zooniverse.py#L270-L365

You can see the resulting subjects here: https://frontend.preview.zooniverse.org/projects/mrniaboc/superluminous-supernovae/classify/workflow/16264

You might need to reload a couple of times to get a working data subject, but they look like this:

ZTF data for a supernova candidate, shown next to a question task that asks if the light curve has been rising for more than 20 days. The subject consists of a plot of magnitude vs. days since first detection, and a panStamps image of the ZTF target.