Minor improvements concerning #6353 and #6357 #6386
Conversation
Extend the description about the normed option for histograms (hist) to clarify which one of the two common ways to normalize the counts is available through this option and which is not.
|
power-cycled to re-run CI against current master (which has had some un-related fixes applied). |
| ``n/(len(x)`dbin)``, i.e., the integral of the histogram will sum | ||
| to 1. If *stacked* is also *True*, the sum of the histograms is | ||
| normalized to 1. | ||
| the integral of the histogram will sum to 1. So, the normalized |
There was a problem hiding this comment.
Can you flip this sentence to say what it is first (before saying what it is not) and make it a bit terser?
Can we assume are audience knows calculus?
There was a problem hiding this comment.
Ok, Thomas, no problem flipping it. Anyway, these two normalization methods of histograms are somehow like miles and kilometers, in the sense that you can say 'Hey, this is not in kilometers, this is in miles'. I can also switch 'the integral of the histogram will sum to 1' to 'the area (or integral) under the histogram will sum to one'.
There was a problem hiding this comment.
I hope it is right now from your point of view. I flipped and shortened the explanation. I think it is in the limit to be useful to avoid new confusion about the normalization method that is applied. Thanks.
Tries to introduce the suggestions of Thomas in #6386.
|
The remaining issue on Travis is a whitespace issue from pep8 /home/travis/build/matplotlib/matplotlib/lib/matplotlib/axes/_axes.py:5807:73: W291 trailing whitespace |
Correct pep8 deviation in _axes.py:5807:73: W291 trailing whitespace
|
Many thanks @jenshnielsen! It should be right now. Let us see the result of the tests... |
|
Thanks! |
|
@tacaswell, @jenshnielsen: Sorry, I know I made this a bit painful! 😌 Thank you for your patience and help! 😉 |
|
Thank you for being patient with us. |
Related with #6353 and #6357, this commit corrects ylabel string in histogram_demo_features.py example but also tries to improve the documentation of hist in pyplot to clarify which one of the two common ways to normalize the counts is available through this option and which is not, in order to avoid future misunderstandings.