The goal of altair is to help you build Vega-Lite visualizations. Using the reticulate package, it provides an interface to the Altair Python package. This approach is distinct from efforts to build a native R interface to Vega-Lite, by Bob Rudis and coworkers: the vegalite package, which has inspired this effort.

In this documentation, the capitalized word Altair shall refer to the Python package; the lower-case word altair shall refer to this R package.

Installation

You can install altair from CRAN with:

The development version of is available from GitHub:

Because you are using a Python package, you may have some additional installation steps. These steps are described in greater detail in the Installation article.

  1. Python must be installed on your system. We have has success using Conda; Miniconda works well and installs more-quickly than Anaconda.

    If you work in a corporate or institutional environment, you may have to specify the location of your SSL certificate, or deal with a proxy. The installation article has a section on this.

  2. Create a Conda environment called "r-reticulate". The reticulate folks recommend standardizing on a common name for all packages that use reticulate. For more information, there is a section in the installation article.

  3. Install Altair into your "r-reticulate" environment using altair::install_altair().

You may wish to add a line like this to the .First() function in your .Rprofile:

reticulate::use_condaenv("r-reticulate")

The use_condaenv() function is called to provide a hint to reticulate on which Python environment to use.

Optional installations

If you have nodejs installed on your computer, you can use vegawidget’s image-generating functions to convert Altair charts into SVG strings or write SVG files. With the rsvg and png packages installed, you can get a bitmap array, or write PNG files.

When knitting to a non-HTML format, e.g. github_document, this package provides a knit_print() function that will intercept the normal renderer, using instead its own renderer, allowing you to specify "png" , "svg" or "pdf". Like the image functions, this requires that nodejs be installed on your system, as well as rsvg and png. MacOS users will require an X11 system, such as XQuartz, to be installed.

Example

Some things to keep in mind:

  • Where you see a . in the Python examples, use a $ instead.

  • In your data, columns that contain dots can be wrapped in square brackets in Altair specifications, e.g. [Sepal.Width], to keep Altair from throwing an error. Alternatively, you can use a double-backslash, e.g. Sepal\\.Width.

These and other “gotchas” are compiled, along with workarounds, in an article: Field Guide to Python Issues.

Articles

The documentation for this package includes some articles:

Gallery:

  • An adaptation to R of the Altair Example Gallery, to demonstrate (not least to ourselves) that we are not missing any of the expressiveness of the Python API. You may be interested in the Interactive Charts examples.

Examples:

Field Guides:

Acknowledgements

This package rests on these foundations:

  • Altair: Python interface to Vega-Lite

  • reticulate: R framework to work with Python

  • Vega-Lite: a grammar of interactive graphics

  • htmlwidgets: R framework to work with JavaScript visualizations

As well, a particular debt is owed to the folks behind the vegalite package, as it provided a lot of the inspiration for these efforts.

This project is a collaborative effort. In addition to the principal authors:

  • Alicia Schep has contributed the concatenation functions, as well as sorted out many of the Python, JavaScript, and package-API issues.

  • Heike Hofmann has been an invaluable advisor, providing incisive feedback, and insight into the fundamentals of interactive graphics.

Contributing

Please note that this project is released with a Contributor Code of Conduct. By participating in this project you agree to abide by its terms.

This project also has a Contributing Guide.