bqplot

0.12.45 · active · verified Mon Apr 13

bqplot is a 2-D visualization system for Jupyter, based on the constructs of the Grammar of Graphics. It provides interactive plotting for Jupyter notebooks, leveraging d3.js and ipywidgets. Every component of a bqplot plot is an interactive widget, allowing seamless integration with other Jupyter interactive widgets to create integrated GUIs. It offers both a flexible object model API and a concise pyplot-like API similar to Matplotlib. The library is actively maintained with frequent patch releases, and a major version (0.13.0) is in pre-release, indicating ongoing development.

Warnings

breaking The `bqplot` 0.12.x series (including 0.12.45) is explicitly compatible with `ipywidgets < 8.0`. Upgrading `ipywidgets` to version 8.0 or higher while using `bqplot` 0.12.x will lead to broken plots and runtime errors.
Fix: For bqplot 0.12.x, ensure `ipywidgets` is installed as `pip install 'ipywidgets>=7,<8'`. To use `ipywidgets 8+`, you will need to upgrade to bqplot 0.13.x (when released, currently in pre-release) or newer.
gotcha For users of JupyterLab <= 2, an additional manual installation of JupyterLab extensions is often required for `bqplot` to render correctly. Without these, plots may appear as text representations (e.g., `<bqplot.figure.Figure object at 0x...>)` instead of interactive visualizations.
Fix: After `pip install bqplot`, also run `jupyter labextension install @jupyter-widgets/jupyterlab-manager bqplot` and restart JupyterLab. This step is typically not needed for JupyterLab >=3 with `ipywidgets 7.6+`.
deprecated In the upcoming `bqplot` 0.13.x release, WebGL-based marks (e.g., `ScatterGL`) are being moved to a separate `bqplot-gl` package. Direct imports of these from `bqplot` will likely be deprecated or removed, requiring explicit imports from `bqplot_gl` and its installation.
Fix: For WebGL marks, install `bqplot-gl` (`pip install bqplot-gl`) and import marks like `ScatterGL` from `bqplot_gl` directly.

Install

pip install bqplot PyPI (recommended)
conda install -c conda-forge bqplot Conda

Imports

pyplot
```
from bqplot import pyplot as plt
```
The Matplotlib-inspired, context-based API for quick plotting.
Figure
```
from bqplot import Figure, LinearScale, Axis, Lines
```
For the object model API, where components are explicitly constructed.

Quickstart

This quickstart demonstrates creating a simple line chart using bqplot's pyplot API, which is similar to Matplotlib's `pyplot`. It generates random data, creates a figure, plots the data, and displays the interactive chart in a Jupyter environment.

import numpy as np
from bqplot import pyplot as plt

# Generate some data
size = 100
np.random.seed(0)
x_data = np.arange(size)
y_data = np.cumsum(np.random.randn(size) * 100.0)

# Create a figure and plot
fig = plt.figure(title="My First Plot")
line_plot = plt.plot(x_data, y_data)

# Display the plot (in a Jupyter environment, this might be implicit or use display(fig))
# For explicit display, you'd use from IPython.display import display; display(fig)
fig

view raw JSON →