Repository navigation
ImageGrid bug when using inline backend #12370
Description
Bug report
Bug summary
The ImageGrid class seems to not behave correctly since the upgrade to matplotlib 3.0.0 when the inline backend is used in a Jupyter notebook: the resulting subplots fail to arrange in a grid and are placed on top of eachother. The correct behaviour is reproduced if using the notebook or ipympl backends. Thanks!
Code for reproduction
import matplotlib.pyplot as plt
from mpl_toolkits.axes_grid1 import ImageGrid
import numpy as np
im = np.arange(100)
im.shape = 10, 10
fig = plt.figure(1, (4., 4.))
grid = ImageGrid(fig, 111,
nrows_ncols=(2, 2),
axes_pad=0.1,
)
for i in range(4):
grid[i].imshow(im)
grid[i].set_title('test {0}{0}'.format(i))
plt.show()Actual outcome
Expected outcome
Matplotlib version
- Operating system: Debian 9
- Matplotlib version: 3.0.0
- Matplotlib backend: inline
- Python version: 3.5.3
- Jupyter version (if applicable): 4.4.0
- Other libraries:
Activity
jklymak commented on Oct 2, 2018
Yeah this seems to be a duplicate of #12355; inline uses bbox_inches='tight', and there was a bad interaction with axes_grid1 subclasses and that saving option. #12363 should fix this as well....
jklymak commented on Oct 2, 2018
Just add fig.savefig('Boo.png', bbox_inches='tight') to the above to make it reproducible w/o a notebook. #12363 indeed fixes this...
jklymak commented on Oct 2, 2018
Thanks for the bug report - closing as a duplicate, but I used your example as a new test...
ezatterin commented on Oct 2, 2018
Understood. Thanks to you!
ImportanceOfBeingErnest commented on Oct 2, 2018
@ezatterin While waiting for the new release which hopefully has this fixed you may always decide to not crop the images when using %matplotlib inline via
%config InlineBackend.print_figure_kwargs = {'bbox_inches':None}in IPython or Jupyter.