How to autohide the Bokeh toolbar when using holoviews

The holoviews documentation mentions that the Bokeh toolbar can be hidden or moved: http://holoviews.org/user_guide/Plotting_with_Bokeh.html The bokeh document shows how to autohide the toolbar so it only shows on mouse-over.

Is it possible to autohide the toolbar when using holoviews as it doesn't allow me to pass options like toolbar='autohide'

Any help would be most welcome.

fundYear.hvplot.bar(
    x='year',
    y='fundingReq',
    rot=90,
).opts(
    toolbar='left',
    title="Funding Requested per Year",
    yformatter='$%f',
)

Solution

Possible settings for the position of the toolbar are:

['above', 'below', 'left', 'right', 'disable', None]

So you can't set autohide like that, but...

1) You can use hooks to set autohide.

With hooks you can customize the plot just before it will be plotted.

def set_toolbar_autohide(plot, element):
    bokeh_plot = plot.state
    bokeh_plot.toolbar.autohide = True

your_plot.opts(hooks=[set_toolbar_autohide], backend='bokeh')

You can also find useful info on hooks in the FAQ:
https://holoviews.org/FAQ.html

2) Another solution would be to convert your Holoviews plot to an actual bokeh plot and then set the bokeh toolbar to autohide:

Quick solution is basically:

my_bokeh_plot = hv.render(my_hv_plot, backend='bokeh')
my_bokeh_plot.toolbar.autohide = True

Full working example of 2nd solution:

# import libraries
import numpy as np
import pandas as pd

import holoviews as hv
import hvplot.pandas
hv.extension('bokeh', logo=False)

from bokeh.plotting import show


# create sample dataframe
df = pd.DataFrame({
    'col1': np.random.normal(size=30),
    'col2': np.random.normal(size=30),
})

# create holoviews plot
my_hv_plot = df.hvplot.scatter(label='Scattering around', width=700)

# turn plot into bokeh plot
my_bokeh_plot = hv.render(my_hv_plot, backend='bokeh')

# set toolbar to autohide
my_bokeh_plot.toolbar.autohide = True

# show plot
show(my_bokeh_plot)