Add sharex and sharey to grouped boxplot #8008
Conversation
sharex and sharey should be settable and by default False in a grouped boxplot. E.g.
In [97]: df = DataFrame(np.random.rand(10,2), columns=['Col1', 'Col2'] )
In [98]: df['X'] =['A','A','A','A','A','B','B','B','B','B']
In [99]: df.iloc[0:5, :2] += 10000
In [100]: df.groupby('X').boxplot() # this is unreadable
| @@ -2732,7 +2734,7 @@ def boxplot_frame_groupby(grouped, subplots=True, column=None, fontsize=None, | |||
| naxes = len(grouped) | |||
| nrows, ncols = _get_layout(naxes, layout=layout) | |||
| fig, axes = _subplots(nrows=nrows, ncols=ncols, naxes=naxes, squeeze=False, | |||
| ax=ax, sharex=False, sharey=True, figsize=figsize) | |||
There was a problem hiding this comment.
This does change the default ... (not to would be to have sharex default to False and sharey default to True). But I changed this to match DataFrame.hist and it makes more sense anyways. Let me know if I should change it to not modify the current default.
|
@danfrankj this needs tests. ok with changing to match |
|
@jreback not exactly sure what a unit test for this kind of option would look like. |
|
take a look here: https://github.com/pydata/pandas/blob/master/pandas/tests/test_graphics.py you need to test for the correctness, e.g. you say sharey=..., the returned axes then should be shared, etc. |
|
closing pls reopen if/when updated |
|
|
sharex and sharey should be settable and by default False in a grouped boxplot. E.g.