You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I can't reproduce the problem; with matplotlib 1.4.0, Anaconda 2.1.0, on OSX, if I cut and paste the code on that page as instructed, it works perfectly.
Actually, the %matplotlib agg translates in to python code, so I could imagine it could be useful for someone doing plain python. Notebooks are just python in the end. I also don't think this is ipython vs. python related, I think it's related to the backend of matplotlib. So my suggestion was to temporarily switch backends.
Moreover, the bug was closed because it couldn't be reproduced. So anyone who googles this error and gets to this page, now at least has a lead.
Activity
pelson commentedon Jan 9, 2015
Are you loading a pickle file created with the same version of matplotlib? Can you provide the code to reproduce. Thanks.
sektoid0 commentedon Jan 9, 2015
Yes. The same version of matplotlib. The code is exactly the one listed here:
http://stackoverflow.com/questions/7290370/store-and-reload-matplotlib-pyplot-object
blink1073 commentedon Jan 22, 2015
We can pickle that
efiring commentedon Feb 4, 2015
I can't reproduce the problem; with matplotlib 1.4.0, Anaconda 2.1.0, on OSX, if I cut and paste the code on that page as instructed, it works perfectly.
prhbrt commentedon Jan 8, 2020
I solved this issue in Jupyter notebooks, not Anaconda, using
That is, this is ipython code, not python code.
QuLogic commentedon Jan 8, 2020
This bug report doesn't involve notebooks. It's also 4 years old and closed. Your workaround may be helpful to you, but it's not really relevant here.
prhbrt commentedon Jan 9, 2020
Actually, the
%matplotlib aggtranslates in to python code, so I could imagine it could be useful for someone doing plain python. Notebooks are just python in the end. I also don't think this is ipython vs. python related, I think it's related to the backend of matplotlib. So my suggestion was to temporarily switch backends.Moreover, the bug was closed because it couldn't be reproduced. So anyone who googles this error and gets to this page, now at least has a lead.