PsychoinformaticsLab / pliers

Automated feature extraction in Python

https://pliers.readthedocs.io/en/latest/

BSD 3-Clause "New" or "Revised" License

296 stars 68 forks source link

Update docs #445

Closed adelavega closed 3 years ago

adelavega commented 3 years ago

Closes #402

I updated the docs to reflect new features added in the last few releases. I didn't make any other substantive changes to the docs, a maintenance update.

In a previous commit (already in master), I added GH Actions to build docs & upload as an artifact for PRs, and upload to gh-pages when merged to master.

codecov-io commented 3 years ago

Codecov Report

Merging #445 (95aff5a) into master (244d08f) will not change coverage. The diff coverage is n/a.

@@           Coverage Diff           @@
##           master     #445   +/-   ##
=======================================
  Coverage   75.73%   75.73%           
=======================================
  Files          65       65           
  Lines        3820     3820           
=======================================
  Hits         2893     2893           
  Misses        927      927

Impacted Files	Coverage Δ
pliers/extractors/api/clarifai.py	`23.01% <ø> (ø)`
pliers/extractors/text.py	`97.00% <ø> (ø)`

Continue to review full report at Codecov.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update 244d08f...95aff5a. Read the comment docs.

adelavega commented 3 years ago

@rbroc do you mind taking a look at the docs to see if theres anything you want to add or improve? just wanted to get the ball rolling.

rbroc commented 3 years ago

had a look and pushed a couple of fixes to typos/issues with docstrings. I think we may still be missing:

AudioResamplingFilter in the Filters section.
All the misc-type stuff (SeriesStim, MetricExtractor, ExtractorResultToSeriesConverter). The converter could deserve a little paragraph of its own (either in the converters section, or where Result objects are discussed). Not sure what _replacements.rst does, but a few extractors are missing there too.

Re: tutorials and quickstart (though not relevant to this PR), probably worth developing a section for Bert, AudioSet and other DL models where complexity scales a bit. Probably also an example with MetricExtractor would be beneficial.

Let's touch base on these points later, happy to help with fixes to docs and/or tutorials. :)

adelavega commented 3 years ago

Yes, I think that's what's missing! Feel free to take a stab if you have some free cycles.

adelavega commented 3 years ago

@rbroc have you had any time to work on the changes you wanted to make?

rbroc commented 3 years ago

done now - maybe do take a quick look just to make sure I did it right ;)

adelavega commented 3 years ago

Roberta, looks good, however now I'm realizing something is going wrong w/ Sphinx. The substitutions are failing at sometimes that I wouldn't expect them to.

Also, the autodefinition tables are blank.

Take a look at the artifact: https://github.com/PsychoinformaticsLab/pliers/suites/2267564946/artifacts/47272079

I'm not sure what's going on, but some of this stems from the prior PR it seems, so perhaps its a more general issue with the GH Actions.

Anyone have any clues? @tyarkoni ?

adelavega commented 3 years ago

See how the autodoc is yielding empty docstrings: http://psychoinformaticslab.github.io/pliers/reference.html#module-pliers.extractors

adelavega commented 3 years ago

Okay, that was a pain but I fixed it. Sphinx could not find pliers. had to remake the entire action w/ custom code but now we're not relying on an obscure github action that wasn't doing much in the first place.

still a few more minor sphinx errors to be fixed but otherwise, this should be close