Classify better

[acjbrouwer]

I have given classifying the datasets a first-order stab. But it can probably be done better given the requisite background knowledge which you @EstherBoere so happen to be in possession of. Hence you have become the lucky assignee of this ticket which involves classifying the datasets better by editing this Python file.

This will be an ongoing task until all datasets have trickled in. It involves binning (putting the right numbers in the right class lists) and, if you care about the order in which the datasets appear in the website, re-ordering the ID numbers in those lists. The syntax is Python dictionary and list syntax, but from just looking at the file, I think it is pretty much self-explanatory. Yes, the last items can have a trailing comma unlike in most other languages.

Once you have modified the file, the website won't update automatically as I will need to run a manual generation step: the file is input to the script that generates the restructured text, at least for now, might turn it into a GitHub Action in the future.

[EstherBoere]

Regrouped databases in the classes but leaving the ticket open as I need to return to it

[acjbrouwer]

classes.py would benefit from an update.

[EstherBoere]

Committed an update to classify.py based on the 13 records listed in trial_zenodo_API.

[acjbrouwer]

Site regenerated with new classification.

[acjbrouwer]

If no more datasets are expected, this issue can be closed.

[EstherBoere]

I do still expect some, will nudge the partners responsible again.

[EstherBoere]

I added 3 more in the classify.py. @acjbrouwer could you run the script again so that they hopefully enter the website?

[acjbrouwer]

Regenerated, if those were the last datasets, please close this issue.

[EstherBoere]

No, still waiting for more, sorry

[acjbrouwer]

Regenerated for new dataset: https://iiasa.github.io/COACCH/en/master/datasets/5744283.html