A Jupyter widget for Vega 5 and Vega-Lite 4

pip install vega==4.0.0


IPython Vega

PyPI npm version Build Status

IPython/Jupyter notebook module for Vega 5, and Vega-Lite 4. Notebooks with embedded visualizations can be viewed on GitHub and nbviewer. If you use JupyterLab (not the notebook), you don't need to install this extension since JupyterLab comes with built-in support for Vega and Vega-Lite.

Available on pypi and Conda Forge as vega.

Install and run

Python Package Index

To install vega and its dependencies from the Python Package Index using pip, use the following commands:

pip install jupyter pandas vega
pip install --upgrade notebook  # need jupyter_client >= 4.2 for sys-prefix below
jupyter nbextension install --sys-prefix --py vega
jupyter nbextension enable --py --sys-prefix vega

Conda Forge

If you use Conda, you probably already have the latest versions of the notebook and pandas installed. To install vega extension run:

conda install vega


Once the package is installed, run

jupyter notebook

to launch the Jupyter notebook server, and use vega within the notebook. See the example notebooks for Vega-Lite and Vega.

To run the notebooks yourself, you need to get the file cars.json.


This project uses Poetry. If you prefer a local virtual environment, run poetry config virtualenvs.in-project true first. Install requirements: poetry install.

Then activate the virtual environment with poetry shell.

Symlink files instead of copying files:

jupyter nbextension install --py --symlink --sys-prefix vega
jupyter nbextension enable --py --sys-prefix vega

Run kernel with jupyter notebook. Run the tests with pytest vega.

To rebuild the JavaScript continuously, run yarn watch.

How to make a release

  • Update the JavaScript dependencies by changing package.json (e.g. with ncu).
  • Run yarn.
  • Update the version number in pyproject.toml (with poetry version [VERSION]), package.json, _frontend.py, and __init__.py
  • Rebuild the JavaScript with yarn build.
  • Make sure that everything still works (launch notebook and widgets and try the examples).
  • Add a git tag.
  • git push --tags.
  • Run npm publish to update https://www.npmjs.com/package/jupyter-vega.
  • Then run poetry publish --build to update https://pypi.python.org/pypi/vega.

The Conda feedstock for this package is at https://github.com/conda-forge/vega-feedstock. It should update automatically but we may need to merge a pull request with the updates.

Visual Regression Tests

ipyvega uses the same technical solution as ipywidgets for visual regression testing (i.e. Galata).

Therefore, the instructions provided for ipywidgets visual regression tests apply here.


  • ipyvega uses ui-tests/tests/notebooks/vega.ipynb notebook for testing
  • reference images are in the ui-tests/tests/vega.test.ts-snapshots/ directory.