Plotting Time Series Using Bokeh

Question

I am working on some Stock (Time-Series) data. I am using Bokeh to visualize the said data.

I was trying to visualize Closing stock Price (stored in a column named Close ). Now there's another column Bool which has boolean data (0 or 1) based on certain calculations I did on the data.

I have to plot the Closing stock such that the colour in a line plot that alters colours between two values of 1 in the Bool column. Here is a real example of some values of the Bool column.

0
1
0
0
1
1
0
1
1
0
0
1

So the Close plot will be blue except for the 0's between two 1's in the Bool column. For those cases, the plot has to be red.

I am not very used to using Bokeh so it would be great if you can help me out here

Answer 1

Did you mean something like this? (works for Bokeh v1.0.4)

import pandas as pd
import numpy as np
from bokeh.models import ColumnDataSource
from bokeh.plotting import figure, show

bools = [0, 1, 0, 0, 1, 1, 0, 1, 1, 0, 0, 1]
colors = ['red' if bool == 0 else 'blue' for bool in bools]

df = pd.DataFrame(data = np.random.rand(len(bools), 1), columns = ['close_price'], index = pd.date_range(start = '01-01-2015', periods = len(bools), freq = 'd'))

xs, ys = [], []
for xs_value, ys_value in zip(df.index.values, df['close_price'].values):
    if len(xs) > 0 and len(ys) > 0:
        xs.append([xs[-1][1], xs_value])
        ys.append([ys[-1][1], ys_value])
    else:
        xs.append([xs_value, xs_value])
        ys.append([ys_value, ys_value])

source = ColumnDataSource(dict(xs = xs, ys = ys, color = colors))

p = figure(width = 500, height = 300, x_axis_type = "datetime")
p.multi_line(xs = 'xs',
             ys = 'ys',
             color = 'color',
             source = source,
             line_width = 3)
show(p)

Result: