Introduction¶

The notebook comes alive with the interactive widgets

Speeding up the bottleneck in the REPL¶

In [ ]:

9*9

In [ ]:

def f(x):
    print(x * x)

In [ ]:

f(9)

In [ ]:

from ipywidgets import *

In [ ]:

interact(f, x=(0, 100));

Interactive Jupyter widgets¶

A Python widget is an object that represents a control on the front end, like a slider. A single control can be displayed multiple times - they all represent the same python object.

In [ ]:

slider = FloatSlider(
    value=7.5,
    min=5.0,
    max=10.0,
    step=0.1,
    description='Input:',
)

slider

In [ ]:

slider

The control attributes, like its value, are automatically synced between the frontend and the kernel.

In [ ]:

slider.value

In [ ]:

slider.value = 8

You can trigger actions in the kernel when a control value changes by "observing" the value. Here we set a global variable when the slider value changes.

In [ ]:

square = slider.value * slider.value
def handle_change(change):
    global square
    square = change.new * change.new
slider.observe(handle_change, 'value')

In [ ]:

square

You can link control attributes and lay them out together.

In [ ]:

text = FloatText(description='Value')
link((slider, 'value'), (text, 'value'))
VBox([slider, text])

Jupyter widgets as a framework¶

Jupyter widgets forms a framework for representing python objects interactively. Some large open-source interactive controls based on Jupyter widgets include:

bqplot - 2d plotting library
pythreejs - low-level 3d graphics library
ipyvolume - 3d plotting and volume rendering
ipyleaflet - interactive maps
ipywebrtc - video streaming
ipysheet - interactive spreadsheets
...