plot_hdi raise exception when x is string (#2412)

[milesalanmoore]

Description

It does not seem that categorical (str) type x values are supported by az.plot_hdi but the documentation for the function is not clear here and no exception is raised if x is type str (i.e. in the case of an ANOVA type model). Here, I've contributed the following:

• A TypeError if x is a str to plot_hdi
• Added test to test_plots_matplotlib.py following the example of plot_hdi: add exception if x is type np.datetime64 and smooth=True #2016.

For now, I prefix the PR with [WIP] to invite input on whether or not to include this lack of support for str type x in the function documentation -- and because this is my first go at contributing to a repository of this caliber...

Fixes #2412

Checklist

• Follows official PR format
• Includes new or updated tests to cover the new feature
• Code style correct (follows pylint and black guidelines)
• Changes are listed in changelog

History

• 2025-01-19 21:18: My own test was failing, I had misplaced the raise statement. Relocated & amended the commit with force-push to keep a concise history.

---

📚 Documentation preview 📚: https://arviz--2413.org.readthedocs.build/en/2413/

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 86.75%. Comparing base (0fc1117) to head (c40c641).
Report is 2 commits behind head on main.

Additional details and impacted files

@@             Coverage Diff             @@
##             main    #2413       +/-   ##
===========================================
+ Coverage   68.26%   86.75%   +18.48%     
===========================================
  Files         124      124               
  Lines       12936    12938        +2     
===========================================
+ Hits         8831    11224     +2393     
+ Misses       4105     1714     -2391     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[amaloney]

thansk @milesalanmoore! I will try to take a look at this later this week

[milesalanmoore]

Thank you, @amaloney, for your time and very thoughtful message (#2412 (comment)). I have modified the error type and message per your suggestions! Let me know what you think. I took a bit of a liberty in the error message and add a recommendation to use az.plot_forest -- but I am happy to remove this or further modify the error message.

[amaloney]

rad, thanks again for this

[milesalanmoore]

Of course, I enjoyed contributing. Thank you for your time!

Cheers!