minor bug fixes #314
minor bug fixes #314
Conversation
Codecov ReportPatch coverage:
Additional details and impacted files@@ Coverage Diff @@
## master #314 +/- ##
=========================================
Coverage 100.00% 100.00%
=========================================
Files 33 33
Lines 2792 2807 +15
=========================================
+ Hits 2792 2807 +15
☔ View full report in Codecov by Sentry. |
…"Frequency" messing with our plotting grid
… in a "Frequency" label messing with our plotting grid
… of the potentially different length of the tick labels (significantly improving appearance in my opinion)
|
To deal with the irritating frequency labelling, it's better to oblate the offending Frequency code. This therefore covers the case when I've got another bug I'd like to squash associated with hist_1d and its plot range (which is why I'm a bit more into this than one might expect) |
| @pytest.mark.parametrize('kind', ['kde', 'hist', 'fastkde'] | ||
| if 'fastkde' in sys.modules else | ||
| ['kde', 'hist']) |
There was a problem hiding this comment.
Is this the best way to deal with fastkde, or should it be skipped/xfailed?
There was a problem hiding this comment.
This achieves the same as skipping. Skipping would appear in the tests as a little s rather than in the form of fewer tests. Would we prefer that?
There was a problem hiding this comment.
Fewer lines in the code is good. Being explicit that tests are skipped when not everything is installed is better. Mentioned in #110.
|
Feel free to squash and merge @lukashergt if you agree with the changes |
neffmethod ofWeighedDataFramewas missing its optionalbetakwarg. Failing test in 1f479b7 highlights the issue.samples.plot_1d()pandas annoyingly introduces the ylabel "Frequency" (see for example the 2.0.0 version of readthedocs), which this PR removes.Checklist:
flake8 anesthetic tests)pydocstyle --convention=numpy anesthetic)python -m pytest)