From the Vega Example Gallery
A heatmap of annual temperatures in Seattle, organized by day versus hour. How do different choices of color palette affect the clarity of the patterns?
Density estimates for automobile statistics, partitioned by region of origin. The density grids are rendered as images by the heatmap transform. By default each heatmap is normalized independently. If resolve equals `shared`, the heatmaps instead show probability or count densities normalized by the global maximum across plots.
Parallel coordinates visualize multi-dimensional data by representing each dimension as a parallel axis, and drawing individual data records as lines connecting points on each axis. Line crossings indicate negative correlation, and different axis orderings may reveal different patterns of interest.
A word cloud (or tag cloud) depicts text data, typically by sizing each word proportionally to its frequency within the text. Vega’s wordcloud transform uses a layout similar to Wordle, based on Jason Davies’ wordcloud implementation. This example visualizes frequent words in the abstracts of published research papers from the Vega project.
A custom timeline visualization showing the lifespans of the first five U.S. presidents, including the years each held office. The timeline is additionally annotated with selected historical events.
A beeswarm plot conveys the size of a group of items by visually clustering the each individual data point. This example uses Vega’s force transform to calculate the clustered layout. The example uses non-standard xfocus and yfocus encoding channels to create anchor coordinates that parameterize the x and y forces.
A calendar visualization of daily changes to the S&P 500 since 2000. Adapted from a Calendar View by Mike Bostock.