BUG: fix handling of color argument for variety of plotting functions #6956
Conversation
| @@ -438,20 +439,23 @@ def normalize(series): | |||
| return ax | |||
|
|
|||
|
|
|||
| def andrews_curves(data, class_column, ax=None, samples=200, colormap=None, | |||
| **kwds): | |||
| def andrews_curves(frame, class_column, ax=None, samples=200, color=None, | |||
There was a problem hiding this comment.
Can you check if the deprecate_kwarg decorator works for the change from data to frame? It may not since its a positional argument. If it doesn't you should change frame back to data since this would break code using andrews_curves(data=df, class_column='Name').
|
Thanks for the PR. I had those couple comments inline. Could you also add the note you have in Does this close a particular Github issue? I'll add this deprecation to #6581 when we merge. |
|
Also it doesn't look like Travis ran. You may need to hook it up to your Github account: http://about.travis-ci.org/docs/user/getting-started/ You should just have to sign in. |
|
I made the changes you suggested. A Travis build seemed to get launched once I pushed my commit but there we errors that seem unrelated to my changes. Do you have any suggestions on what might be the issue? Here is the build https://travis-ci.org/anomrake/pandas/builds/24206970 |
|
pls rebase to master. travis made some changes over the last couple of days, all fixed now. travis will automatically build on each commit. |
|
I tried to rebase but I'm new to this and I'm a bit surprised to see these other commits show up. Any help would be great. |
|
see here: https://github.com/pydata/pandas/wiki/Using-Git then |
|
Looks to have worked now. Thanks for the help. Also, as far as I know, there was no Github issue for the bugs. |
| @@ -464,6 +464,14 @@ Deprecations | |||
| returned if possible, otherwise a copy will be made. Previously the user could think that ``copy=False`` would | |||
| ALWAYS return a view. (:issue:`6894`) | |||
|
|
|||
| - The :func:`parallel_coordinates` function now takes argument ``color`` | |||
| instead of ``colors``. A ``FutureWarning`` is raised to alert that | |||
There was a problem hiding this comment.
pls move these to the Plotting sub-section
|
I made the change you requested |
| if ax is None: | ||
| ax = plt.gca(xlim=(-pi, pi)) | ||
| for i in range(n): | ||
| row = [columns[c][i] for c in range(len(columns))] | ||
| row = df.irow(i).values |
There was a problem hiding this comment.
Sorry I missed this the first time. I think we're deprecating irow and icol sometime. Could you change this to df.iloc[i].values
|
If you could rebase one last time and change those |
parallel_coordinates - fix reordering of class column (from set) causing possible color/class mismatch - deprecated use of argument colors in favor of color radviz - fix reordering of class column (from set) causing possible color/class mismatch - added explicit color keyword argument (avoids multiple values 'color' being passed to plotting method) andrews_curves - added explicit color keyword argument (avoids multiple values 'color' being passed to plotting method)
parallel_coordinates/andrews_curves - added deprecate_kwarg decorator for using frame argument instead of data - added tests to check that FutureWarning is raised properly
|
OK done |
BUG: fix handling of color argument for variety of plotting functions
|
Great, thanks! |
parallel_coordinates
mismatch
radviz
mismatch
passed to plotting method)
andrews_curves
passed to plotting method)
To recreate the bug:
parallel_coordinates before:
parallel_coordinates after:
