transition.RmdIf you’re familiar with ggplot2 this vignette provides a quick guide to getting up to speed with virgo. The following table provides a mapping between ggplot2 concepts and virgo functions.
| Concept | ggplot2 function | virgo function |
|---|---|---|
| Initialisation | ggplot() |
vega() |
| Aesthetic mappings | aes() |
enc() |
| Layers | geom_* |
mark_* |
| Adding layers | + |
%>% |
| Facets (grid layouts) | facet_grid() |
facet_views() |
| Scales | scale_{aesthetic}_type() |
scale_{encoding}(type = type) |
| Statistical transformations | stat_() |
Performed client side with vg family of functions. |
| Coordinates | coord_* |
Not directly supported |
| View concatenation | See patchwork package |
hconcat(), vconcat()
|
| Interactivity | Not directly supported, but see plotly::ggplotly()
|
select_*() |
As you can see most of the grammar elements in ggplot2 have equivalents in virgo.
Compared to ggplot2, virgo has fewer available layers (one that we wish was available is hex binning) and less support for spatial visualisations.
Unlike ggplot2, the sizing of a virgo visualisation has to be declared in vega(). Here the width and height determine the pixel dimensions on the canvas/Rstudio Viewer. By default, we set these to 300. When faceting plots, virgo the height/width refers to dimensions of the entire view not each subplot.
To save a static visualisation with virgo, use the vegawidget functions vw_write_png() or vw_write_svg().