Add mypy and black to CI

WardBrian commented 1 year ago

As discussed in #16, this will run mypy and black against PRs. At the moment it will fail because main doesn't pass these checks, but after main does I will rebase and merge this

codecov-commenter commented 1 year ago

Codecov Report

Merging #26 (90bcde3) into main (3157c8f) will decrease coverage by 0.09%. The diff coverage is 100.00%.

@@            Coverage Diff             @@
##             main      #26      +/-   ##
==========================================
- Coverage   98.10%   98.01%   -0.09%     
==========================================
  Files          11       11              
  Lines         316      302      -14     
==========================================
- Hits          310      296      -14     
  Misses          6        6

Impacted Files	Coverage Δ
bayes_kit/__init__.py	`100.00% <100.00%> (ø)`
bayes_kit/autocorr.py	`100.00% <100.00%> (ø)`
bayes_kit/ensemble.py	`100.00% <100.00%> (ø)`
bayes_kit/ess.py	`100.00% <100.00%> (ø)`
bayes_kit/hmc.py	`94.59% <100.00%> (-0.28%)`	:arrow_down:
bayes_kit/iat.py	`100.00% <100.00%> (ø)`
bayes_kit/mala.py	`93.54% <100.00%> (-0.40%)`	:arrow_down:
bayes_kit/metropolis.py	`100.00% <100.00%> (ø)`
bayes_kit/rhat.py	`100.00% <100.00%> (ø)`
bayes_kit/smc.py	`95.83% <100.00%> (-0.09%)`	:arrow_down:
... and 1 more

:mega: We’re building smart automated test selection to slash your CI/CD build times. Learn more

WardBrian commented 1 year ago

l 48 warns that the call to scipy.stats.beta.logpdf returns an NDArray not a float

Pylance is lying to you

>>> isinstance(scipy.stats.beta.logpdf(.1, 3, 2), float)
True

(it does return an array if the first argument is an array, but it isn't in this model)

I think I resolved the other pylance warnings, though I'm not sure if they were also spurious

flatironinstitute / bayes-kit

Add mypy and black to CI #26

Codecov Report