optimizing naive_bayes_classifier

[afinit]

Describe your changes

This is to help optimize the naive_bayes_classifier a bit. I think it is a little more readable as well.

The piece of this function that is taking the most time is the calculation of joint probabilities, so i focused on changes in that space. After several iterations, I settled on the approach here where we're grabbing the indices for the PDFs separately from the loops. We are now doing this once per function call instead of for every class and this allows us to use numpy index slicing and array multiplication.

I did all the testing on this using the Naive Bayes Tutorial notebook in the plantcv-binder repo.
After getting all of the changes in, I ran the original code 30x and the new code 30x to get runtimes in milliseconds. The screenshot is pulled from that notebook where we see a speedup of about 4x.

This example contains 4 classes that we're processing here, so I imagine this optimization hangs on number of classes.

Type of update
feature enhancement

Associated issues
#1158

For the reviewer
See this page for instructions on how to review the pull request.

• PR functionality reviewed in a Jupyter Notebook
• All tests pass
• Test coverage remains 100%
• Documentation tested
• New documentation pages added to plantcv/mkdocs.yml
• Changes to function input/output signatures added to updating.md
• Code reviewed
• PR approved

[codecov]

Codecov Report

Merging #1223 (5cad126) into release-4.0 (17d8c37) will not change coverage.
The diff coverage is 100.00%.

@@              Coverage Diff              @@
##           release-4.0     #1223   +/-   ##
=============================================
  Coverage       100.00%   100.00%           
=============================================
  Files              161       161           
  Lines             6957      6952    -5     
=============================================
- Hits              6957      6952    -5     

Flag	Coverage Δ
unittests	100.00% <100.00%> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
plantcv/plantcv/naive_bayes_classifier.py	100.00% <100.00%> (ø)