Using interactive widgets¶

You can use ipywidgets and FigureWidget in Notebooks to bring interactive controls such as sliders, buttons, dropdowns, and text inputs. By connecting Python code to browser-based widgets, you can build dynamic visualizations, parameter exploration, and interactive applications.

Installation¶

Runtime v2.6 and above¶

ipywidgets and anywidget are pre-installed. No extra setup is needed: skip straight to importing.

Runtime v2.5 and below¶

Install the packages at the top of your notebook before importing:

%pip install ipywidgets
%pip install anywidget  # required for FigureWidget and other anywidget-backed libraries

Run those cells once per session. After installation, proceed to the imports below.

Imports¶

import ipywidgets as widgets  # interactive controls
import plotly.graph_objects as go  # FigureWidget (backed by anywidget)

For common helpers used in the examples below:

from IPython.display import display

Basic controls¶

Slider¶

slider = widgets.IntSlider(value=5, min=0, max=10, description='Count:')
display(slider)

Dropdown¶

mode = widgets.Dropdown(
    options=['Standard', 'Enterprise', 'Business Critical'],
    value='Standard',
    description='Edition:',
)
display(mode)

Checkbox¶

enabled = widgets.Checkbox(value=True, description='Enable feature')
display(enabled)

Button¶

btn = widgets.Button(description='Apply', button_style='primary')

def on_click(b):
    print('Clicked!')

btn.on_click(on_click)
display(btn)

Text input¶

name = widgets.Text(value='Snowflake', description='Name:')
display(name)

Laying out widgets¶

Use HBox (horizontal) and VBox (vertical) to arrange controls:

display(widgets.VBox([
    widgets.HBox([slider, mode]),
    widgets.HBox([enabled, btn]),
]))

FigureWidget (Plotly)¶

FigureWidget is a Plotly figure that supports live in-place updates, making it ideal for linking controls to a chart without re-rendering the whole figure.

Prerequisite: anywidget must be installed.

Basic line chart¶

import numpy as np

x = np.linspace(0, 2 * np.pi, 200)
fig = go.FigureWidget(data=[go.Scatter(x=x, y=np.sin(x), mode='lines')])
display(fig)

Supported widget types¶

Notes¶

• Widgets require an active kernel to sync state. If you refresh the page, re-run the cell to restore the widget.
• While another cell is executing, interactive controls (sliders, dropdowns, buttons) are temporarily disabled to prevent state drift between the UI and the running kernel.
• Cache isn’t supported with different sessions.

Known unsupported ipywidgets¶

The following widgets show an “Unsupported widget” message:

• Play (PlayModel): Timer-based auto-increment not implemented.
• Image (ImageModel): Binary buffer support not implemented.
• FileUpload (FileUploadModel): Binary buffer and security review needed.