Classes that have fewer students than what we define as "small" should be merged

[patudom]

Perhaps we want to sent up a flag in the db to mark test classes as such and ignore them

[patudom]

In case it's helpful for diagnosing this issue, the too small classes break the age ordering. (Jumps from 14 Gyr to 13 Gyr back to 14 Gyr.)

(Also, the 13 Gyr jump is only in the age label for the best fit line. The slider label says 14 Gyr).

Screen.Recording.2024-12-18.at.9.27.31.AM.mov

[Carifio24]

I think most of these are classes that were made for some testing purpose and that we can ultimately just delete anyways. I set a few that were obviously test classes to me to have test = 1 in the database, but we should go through once we've finished fixing things up and give the class list a final pass.

[patudom]

Thanks, @Carifio24. Is the fact that the age in the slider & the age in the best fit line legend don't always match unrelated to this issue?

[Carifio24]

I believe it should be unrelated, yeah. I'll take a look at that today and see if I can get those to agree.

[patudom]

Between the above PR, cosmicds/cds-api#168, and https://github.com/cosmicds/update-hubble-summaries/pull/7, the issue with the mis-matched ages and classes with blank viewers has been resolved.

Will leave this issue open, so we remember to give the class list the final pass Jon mentioned above.