ENH: Added 'doane' and 'sqrt' estimators to np.histogram in numpy.function_base #7083
Conversation
Refactored methods that broadcast arguments together by finding additional common ground between code in the if...else branches that involved a size parameter being passed in.
MAINT: Simplify mtrand.pyx helpers
| Doane's Estimator | ||
| Improved version of Sturges' formula which works better for non-normal | ||
| data. | ||
| See http://stats.stackexchange.com/questions/55134/doanes-formula-for-histogram-binning |
There was a problem hiding this comment.
I'd rather refer to a wikipedia page or so. Turns out https://en.wikipedia.org/wiki/Histogram does describe it and give the original reference.
There was a problem hiding this comment.
While I would normally agree with you, the author of the better researched Stack Exchange answer is also the author of the relevant section on Wikipedia. I referenced the Stack Exchange article because it is not only the original source, but also contains a lot more information.
There was a problem hiding this comment.
I added a link to Wikipedia in the overall function documentation, since it is relevant to all the estimators.
MAINT: Update the git .mailmap file.
|
If you look at the commits you are trying to merge, you will notice there is something very wrong. You are trying to merge your changes from your master branch, but should be doing so from a feature branch. You can find detailed instructions here. Not sure if we can salvage this, or if you are better off starting from scratch with a new PR... |
|
That is my mistake. I had thought that I read the docs carefully, but I did not notice that http://docs.scipy.org/doc/numpy-1.10.1/dev/gitwash/development_workflow.html#pushing-changes-to-the-main-repo is only for developers with write access. My work is actually in a feature branch. I will go back and create a new PR. |
|
I went ahead and replaced this request with #7090. All changes have been squashed into a feature branch there. |
These are a couple of estimators I find myself using sometimes. They do not break existing code and tests show that they work sensibly.