3 Renderers & Visual Customizations

Python For Data Science

Glyphs Legend Location
Scatter Markers Inside Plot Area
Bokeh Cheat Sheet >>> p1.circle(np.array([1,2,3]), np.array([3,2,1]),

>>> p.legend.location = 'bottom_left'

fill_color='white')
Out side Plot Area
>>> p2.square(np.array([1.5,3.5,5.5]), [1,4,3],

Learn Bokeh online at www.DataCamp.com

color='blue', size=1) >>> from bokeh.models import Legend

>>> r1 = p2.asterisk(np.array([1,2,3]), np.array([3,2,1])

taught by Bryan Van de Ven, core contributor >>> r2 = p2.line([1,2,3,4], [3,4,5,6])

>>> legend = Legend(items=[("One" ,[p1, r1]),("Two",[r2])],

Line Glyphs location=(0, -30))

>>> p1.line([1,2,3,4], [3,4,5,6], line_width=2)

>>> p.add_layout(legend, 'right')
>>> p2.multi_line(pd.DataFrame([[1,2,3],[5,6,7]]),

Plotting With Bokeh pd.DataFrame([[3,4,5],[3,2,1]]),

color="blue") Legend Orientation

The Python interactive visualization library Bokeh enables high-performance
>>> p.legend.orientation = "horizontal"

visual presentation of large datasets in modern web browsers.

Customized Glyphs Also see Data

>>> p.legend.orientation = "vertical"

Bokeh’s mid-level general purpose bokeh.plotting interface is centered Legend Background & Border
around two main components: data and glyphs. Selection and Non-Selection Glyphs
>>> p = figure(tools='box_select')
>>> p.legend.border_line_color = "navy"

>>> p.circle('mpg', 'cyl', source=cds_df,

>>> p.legend.background_fill_color = "white"
selection_color='red',

nonselection_alpha=0.1)
Rows & Columns Layout
Hover Glyphs
>>> from bokeh.models import HoverTool
Rows
>>> hover = HoverTool(tooltips=None, mode='vline')
>>> from bokeh.layouts import row

The basic steps to creating plots with the bokeh.plotting interface are:
>>> p3.add_tools(hover) >>> layout = row(p1,p2,p3)
1. Prepare some data (Python lists, NumPy arrays, Pandas DataFrames and other sequences of values)

Columns
2. Create a new plot

>>> from bokeh.layouts import columns

3. Add renderers for your data, with visual customizations

Colormapping >>> layout = column(p1,p2,p3)
4. Specify where to generate the output
>>> from bokeh.models import CategoricalColorMapper
Nesting Rows & Columns
5. Show or save the results >>> color_mapper = CategoricalColorMapper(

>>>layout = row(column(p1,p2), p3)

factors=['US', 'Asia', 'Europe'],

>>> from bokeh.plotting import figure

palette=['blue', 'red', 'green'])

>>> from bokeh.io import output_file, show

>>> x = [1, 2, 3, 4, 5] #Step 1

>>> p3.circle('mpg', 'cyl', source=cds_df,

color=dict(field='origin',
Grid Layout
>>> y = [6, 7, 2, 4, 5]
transform=color_mapper),

>>> p = figure(title="simple line example", #Step 2

legend='Origin') >>> from bokeh.layouts import gridplot

x_axis_label='x',
>>> row1 = [p1,p2]

y_axis_label='y')
>>> row2 = [p3]

>>> p.line(x, y, legend="Temp.", line_width=2) #Step 3

4 Output & Export

>>> layout = gridplot([[p1,p2],[p3]])
>>> output_file("lines.html") #Step 4

>>> show(p) #Step 5

Tabbed Layout
Notebook
Lists, NumPy & Pandas >>> from bokeh.models.widgets import Panel, Tabs

1 Data Also see

>>> tab1 = Panel(child=p1, title="tab1")

>>> from bokeh.io import output_notebook, show

>>> tab2 = Panel(child=p2, title="tab2")

>>> output_notebook() >>> layout = Tabs(tabs=[tab1, tab2])

Under the hood, your data is converted to Column Data Sources.

You can also do this manually:

HTML Linked Plots
>>> import numpy as np

>>> import pandas as pd

>>> df = pd.DataFrame(np.array([[33.9,4,65, 'US'],

Standalone HTML Linked Axes

[32.4, 4, 66, 'Asia'],
>>> from bokeh.embed import file_html
>>> p2.x_range = p1.x_range

[21.4, 4, 109, 'Europe']]),

>>> from bokeh.resources import CDN
>>> p2.y_range = p1.y_range
columns=['mpg','cyl', 'hp', 'origin'],
>>> html = file_html(p, CDN, "my_plot")

index=['Toyota', 'Fiat', 'Volvo'])

Linked Brushing
>>> from bokeh.io import output_file, show
>>> p4 = figure(plot_width = 100, tools='box_select,lasso_select')

>>> from bokeh.models import ColumnDataSource

>>> output_file('my_bar_chart.html', mode='cdn') >>> p4.circle('mpg', 'cyl', source=cds_df)

>>> cds_df = ColumnDataSource(df) >>> p5 = figure(plot_width = 200, tools='box_select,lasso_select')

Components
>>> p5.circle('mpg', 'hp', source=cds_df)

>>> from bokeh.embed import components

>>> layout = row(p4,p5)
>>> script, div = components(p)
2 Plotting
PN G 5 Show or Save Your Plots
>>> from bokeh.plotting import figure

>>> p1 = figure(plot_width=300, tools='pan,box_zoom')

>>> from bokeh.io import export_png

>>> p2 = figure(plot_width=300, plot_height=300,

>>> show(p1)

>>> export_png(p, filename="plot.png")

x_range=(0, 8), y_range=(0, 8))
>>> show(layout)

>>> p3 = figure() >>> save(p1)

SV G >>> save(layout)

>>> from bokeh.io import export_svgs

>>> p.output_backend = "svg"

>>> export_svgs(p, filename="plot.svg") Learn Data Skills Online at www.DataCamp.com