Closed andgoldschmidt closed 1 year ago
Ok, good point. I'll dig into this. How would you feel about committing the notebook outputs?
Open to it, but definitely appreciated having the automated notebook runs. My current status is: I have no trouble compiling locally when I allow the the notebooks to run. Meanwhile, RTD fails whether I include the notebooks or not:
Extension error:
Could not import extension sphinx.builders.linkcheck (exception: urllib3 v2.0 only supports OpenSSL 1.1.1+, currently the 'ssl' module is compiled with OpenSSL 1.0.2n 7 Dec 2017. See: https://github.com/urllib3/urllib3/issues/2168)
I think https://github.com/readthedocs/readthedocs.org/issues/10290 is perhaps what is needed, will try later.
Good troubleshooting on that issue! On my end I'm getting the error
TypeError: max_iter must be an instance of int, not float.
Whenever example.ipynb sets maxiter=1e6
. So it's likely a version issue w/scikit-learn, but it's an easy fix to cast as int in the notebook. Just running into some issues refreshing the sphinx cache, then I'll PR a fix.
Another desiderata: add a doc build step to the CI to catch any PR that breaks the doc build. I can add this to the PR too. :laughing:
Closed by #34
This should be taken care of now, docs build for me on RTD using the newer .readthedocs.yaml
format.
CI for docs on pull requests has also been added and passes
Current build config does not re-run the example jupyter notebook when building docs (It was causing an error that RTD ignores, see #30). This means the results from running notebook examples are no longer viewable in the docs, so this should be restored / the RTD error addressed or an alternative workflow should be introduced.