Plot Tools

Bokeh plots are interactive by default: the toolbar provides pan, zoom, box-select, lasso, hover, and reset tools. You choose which tools appear, add custom ones, and wire hover tooltips to your data - the core of making a chart feel like a dashboard.

Default Tools

from bokeh.plotting import figure

p = figure(tools="pan,wheel_zoom,box_zoom,reset,hover,save")
# tools can be a comma string or a list of tool instances

from bokeh.plotting import figure
from bokeh.models import Toolbar, HoverTool

p = figure()
p.circle([1, 2, 3], [4, 5, 6])

p.toolbar.logo = "none"          # remove the Bokeh logo
p.toolbar.location = "above"     # above, below, left, right
p.toolbar.button_style = "outline"

Hover Tool and Tooltips

from bokeh.models import HoverTool

hover = HoverTool(
    tooltips=[
        ("index", "$index"),
        ("data (x, y)", "($x, $y)"),
        ("screen", "($xsin, $ysin)"),
        ("name", "@name"),
    ],
)
p = figure(tools=[hover])
p.circle([1, 2, 3], [4, 5, 6])

Special tokens: $x/$y are data coordinates, $xsin/$ysin are screen pixels, @column references a ColumnDataSource field, $name gives the point index.

Selection Tools

p = figure(tools="pan,box_select,lasso_select,undo,redo,reset")
p.circle([1, 2, 3, 4], [1, 2, 3, 4])

# inspect the selection:
p.select(type=BoxSelectTool).dimensions = "width"  # 1-D box selection

Adding and Removing Tools

from bokeh.models import SaveTool, ResetTool

p.add_tool(SaveTool())
p.add_tool(ResetTool())
# remove all tools of a type:
from bokeh.models import PanTool
p.tools = [t for t in p.tools if not isinstance(t, PanTool)]

Disabling Interaction

p = figure(tools="")        # static chart, e.g. for export to a report

Summary

Control plot interactivity with the tools argument (pan, wheel_zoom, box_zoom, reset, hover, save, box_select, lasso_select), configure the toolbar (logo, location), add HoverTool with tooltip templates ($x, @column, $index), and add or remove specific tool instances with add_tool. Set tools="" for a static export.