I'm hoping someone can tell me (point me to the ChanegLog for) when he axes kwarg was removed from matplotlib.pyplot.clabel, matplotlib.pyplot.contou, and matplotlib.pyplot.contourf.
My search has turned up nothing. I have code in an exercise for a class that called: plt.contourf(xx, yy, Z, cmap=ccolor, alpha=.2, axes=ax) cs2 = plt.contour(xx, yy, Z, cmap=ccolor, alpha=.6, axes=ax) plt.clabel(cs2, fmt = '%2.1f', colors = 'k', fontsize=14, axes=ax) when run, this raised errors. I've changed it to ax.contourf(xx, yy, Z, cmap=ccolor, alpha=.2) cs2 = ax.contour(xx, yy, Z, cmap=ccolor, alpha=.6) ax.clabel(cs2, fmt = '%2.1f', colors = 'k', fontsize=14) and the errors are gone. However, a confirming reference would be helpful. _______________________________________________ Matplotlib-users mailing list [hidden email] https://mail.python.org/mailman/listinfo/matplotlib-users |
I don't think it ever was a keyword argument. Now, there have been some bugs fixed recently where unused keyword arguments were going through without emitting an error, which caused confusion when kwargs were misspelled. Perhaps this is the change that is impacting you. To be clear, I don't think it never would have done anything in the past. I hope this helps! Cheers! Ben Root On Fri, Oct 25, 2019 at 9:40 PM Vicki Brown <[hidden email]> wrote:
_______________________________________________ Matplotlib-users mailing list [hidden email] https://mail.python.org/mailman/listinfo/matplotlib-users |
Thanks. That could be what was happening!
-- Vicki Vicki Brown cfcl.com/vlb _______________________________________________ Matplotlib-users mailing list [hidden email] https://mail.python.org/mailman/listinfo/matplotlib-users |
Vicki, https://github.com/matplotlib/matplotlib/pull/9111 is trying to _add_ the ability to pass the axes into the pyplot methods via keyword. From your question, I infer that you would like this functionality to exist? Tom On Sat, Oct 26, 2019 at 12:28 AM Vicki Brown <[hidden email]> wrote:
Thomas Caswell
[hidden email] _______________________________________________ Matplotlib-users mailing list [hidden email] https://mail.python.org/mailman/listinfo/matplotlib-users |
Tha would actually be useful. Many methods have it, others don't. Consistency is good and why have two methods when one will suffice?
- V.
-- Vicki Vicki Brown cfcl.com/vlb _______________________________________________ Matplotlib-users mailing list [hidden email] https://mail.python.org/mailman/listinfo/matplotlib-users |
Just to be clear, no methods have this right now. I am not sure why you think "many methods have it". This is a design proposal. This will also not eliminate any methods, contrary to your "why have two methods when one will suffice". The ultimate goal of the design proposal (at least, how I view it) is to begin to unify how plotting APIs look across packages. Ben Root On Thu, Oct 31, 2019 at 11:50 AM Vicki Brown <[hidden email]> wrote:
_______________________________________________ Matplotlib-users mailing list [hidden email] https://mail.python.org/mailman/listinfo/matplotlib-users |
My apologies. I'm VERY new to this and my wording was imprecise
Many _pandas methods_ have the ax argument. >> why have two methods when one will suffice Why force he user o go find another method when hey've already used one and only need to add one bit. (beter wording? No?) >> The ultimate goal of the design proposal (at least, how I view it) is to begin to unify how plotting APIs look across packages. aka "consistency". (you have an awesome name, btw. ;-)
-- Vicki Vicki Brown cfcl.com/vlb _______________________________________________ Matplotlib-users mailing list [hidden email] https://mail.python.org/mailman/listinfo/matplotlib-users |
Free forum by Nabble | Edit this page |