How to handle labels when using dark theme with jupyterlabs? about ipympl HOT 7 CLOSED

You can use a style sheet, but it looks like styles don't affect plot titles, so those have to be set manually:

%matplotlib ipympl
import matplotlib.pyplot as plt
plt.style.use(['dark_background'])
plt.plot([0,1,12])
plt.suptitle('figure title', color='w')

from ipympl.

1. We can use following for changing background:-
   plt.patch.set_facecolor('lightgrey')
   But this will work only for single plot. Not for grid plot, or more than one.

2. We can use following for getting graphs like we get in jupyter lab light theme:-
   import matplotlib.pyplot as plt
from matplotlib import style
style.use('ggplot') or plt.style.use('ggplot')
   We need to specify "style.use('ggplot')" again in any next cell and not in the cell where we first set the style. Yes, we need to set it 2 times. Second should be the lighter theme.
   

style.available or plt.style.available
['Solarize_Light2',
'grayscale',
'ggplot',
'seaborn-dark-palette',
'seaborn-poster',
'tableau-colorblind10',
'seaborn-ticks',
'seaborn-muted',
'seaborn-bright',
'_classic_test',
'seaborn-darkgrid',
'seaborn-paper',
'seaborn',
'seaborn-colorblind',
'classic',
'seaborn-talk',
'seaborn-notebook',
'seaborn-dark',
'fivethirtyeight',
'fast',
'bmh',
'seaborn-pastel',
'seaborn-deep',
'seaborn-white',
'seaborn-whitegrid'
'dark_background']

lighter ones are:
{0: 'Solarize_Light2', 1: 'grayscale', 2: 'ggplot', 6: 'seaborn-ticks', 9: '_classic_test', 10: 'seaborn-darkgrid', 12: 'seaborn', 14: 'classic', 17: 'seaborn-dark', 18: 'fivethirtyeight', 23: 'seaborn-white', 24: 'seaborn-whitegrid'}

darker ones are:
{3: 'seaborn-dark-palette', 4: 'seaborn-poster', 5: 'tableau-colorblind10', 7: 'seaborn-muted', 8: 'seaborn-bright', 11: 'seaborn-paper', 13: 'seaborn-colorblind', 15: 'seaborn-talk', 16: 'seaborn-notebook', 19: 'fast', 20: 'bmh', 21: 'seaborn-pastel', 22: 'seaborn-deep', 25: 'dark_background'}

from ipympl.

Its no more a problem now...perhaps today onwards.
Just update your concerned python libraries(specially jupyterlab):-
pip install --upgrade numpy pandas matplotlib seaborn jupyterlab

from ipympl.

Yeah the problem isn't the style sheet, it is our rendered title:

from ipympl.

I found one solution for expanding the plot canvas a bit so that x and y ticks will get visible.

You need to add one line plt.style.use('default') after importing the matplot lib module in notebook.

See my ticks are outside the canvas before.

After adding the line the tick got added to the canvas itself no further modification was needed.

As you can see in the above image this will pull the x and y ticks in the canvas to make it visible without changing or modifying ticks or plot color, it will make it easier to collaborate your work with others.

from ipympl.

Actually, we still have a problem with the figure title:

from ipympl.

Closing since the jlab theme cannot set the matplotlib style.

However, in master, the toolbar now uses the JupyterLab CSS variables.

from ipympl.