]> granicus.if.org Git - python/commitdiff
[3.7] bpo-33573: docs to suggest median() alternatives for non-numeric data (GH-7587...
authorMiss Islington (bot) <31488909+miss-islington@users.noreply.github.com>
Mon, 25 Jun 2018 11:18:54 +0000 (04:18 -0700)
committerTal Einat <taleinat+github@gmail.com>
Mon, 25 Jun 2018 11:18:54 +0000 (14:18 +0300)
(cherry picked from commit fdd6e0bf18517c3dc5e24c48fbfe890229fad1b5)

Doc/library/statistics.rst

index bc3817836b9398247e2da94d56672c7088320615..26bb592b23812b8d36447d72820122ff1f18556d 100644 (file)
@@ -169,6 +169,10 @@ However, for reading convenience, most of the examples show sorted sequences.
    This is suited for when your data is discrete, and you don't mind that the
    median may not be an actual data point.
 
+   If your data is ordinal (supports order operations) but not numeric (doesn't
+   support addition), you should use :func:`median_low` or :func:`median_high`
+   instead.
+
    .. seealso:: :func:`median_low`, :func:`median_high`, :func:`median_grouped`