Bug report
Bug summary
When using a combination of plt.subplots with gridspec_kw/width_ratios along with plt.subplot2grid, some of the axes disappear when one of the width ratios is (in this case) below 0.73.
Code for reproduction
import matplotlib.pyplot as plt
fig, ax = plt.subplots(2,3,sharex=True, sharey=True, gridspec_kw={'width_ratios':[1,1,0.7]})
axi = plt.subplot2grid((2,3),(0,2),rowspan=2)
plt.show()
Actual outcome

Note that the axes only disappear with the call to subplot2grid. If I remove that line, all axes appear.
Expected outcome
If width_ratios is instead [1,1,0.8], all of the axes appear.

I can achieve the desired width ratios if I use GridSpec directly instead of using the wrappers, but it would be nice if the wrappers didn't fail based on the width_ratios parameter.
Matplotlib version
- Operating system: Windows 10.0.15063
- Matplotlib version: 2.1.0
- Matplotlib backend: module://ipykernel.pylab.backend_inline
- Python version: 3.6.3
- IPython version: 6.1.0
- Jupyter notebook server version: 5.4.1
mabplotlib and python installed via conda.
Bug report
Bug summary
When using a combination of
plt.subplotswithgridspec_kw/width_ratiosalong withplt.subplot2grid, some of the axes disappear when one of the width ratios is (in this case) below 0.73.Code for reproduction
Actual outcome
Note that the axes only disappear with the call to
subplot2grid. If I remove that line, all axes appear.Expected outcome
If

width_ratiosis instead[1,1,0.8], all of the axes appear.I can achieve the desired width ratios if I use
GridSpecdirectly instead of using the wrappers, but it would be nice if the wrappers didn't fail based on thewidth_ratiosparameter.Matplotlib version
mabplotlib and python installed via conda.