updating pandas-bokeh charts with additional elements

Question

I'm working in a Jupyter Notebook with pandas and pandas-bokeh.

I can create a chart no problem, but when I want to add an element (like a Span), I can't figure out how to avoid getting a second copy of the chart displayed in the notebook.

Is there a way to any of these:

• Delay drawing the figure until I'm ready to show() it?
• Update the existing figure in place?
• Delete the existing figure before drawing the second figure?

This code demonstrates the issue:

# spantest.ipynb

import pandas as pd
import numpy as np
import pandas_bokeh
from bokeh.models import Span
from bokeh.plotting import show

pandas_bokeh.output_notebook()

# Example linechart code from pandas-bokeh docs:
# https://github.com/PatrikHlobil/Pandas-Bokeh#lineplot

np.random.seed(42)
df = pd.DataFrame({"Google": np.random.randn(1000)+0.2, 
                   "Apple": np.random.randn(1000)+0.17}, 
                   index=pd.date_range('1/1/2000', periods=1000))
df = df.cumsum()
df = df + 50
fig = df.plot_bokeh(kind="line")

# So far, so good. At this point, I have one chart displayed.
# Now let's say I want to include a Span to indicate some threshold:

span = Span(location=150, dimension='width', line_color='green')
fig.add_layout(span)
show(fig)

# Now I can see the update, but it is displayed as a second copy of the chart. :-(

Answer 1

Found the answer in the docs (imagine that).

Had to add the option show_figure=False to the figure creation line to prevent it from being shown on figure creation:

fig = df.plot_bokeh(kind="line", show_figure=False)

Then later it appears only once when I call show(fig)

I could swear I tried that at some point before posting the question. Ah well...