In uRosConf/voronoiTreemap: Voronoi Treemaps with Added Interactivity by Shiny

knitr::opts_chunk$set(echo = FALSE)

Idea

• Make it easy to create a plot like this:

knitr::include_graphics("preisKalei.jpg")

R package

• https://github.com/uros2018/voronoiTreemap
• important functions:
  • vt_input_from_df ... easy data input as a data frame
  • vt_export_json ... export to json
  • vt_d3 ... create an htmlwidget
  • vt_app ... start a shiny to create a Voronoi treemap

library(voronoiTreemap)
data(ExampleGDP)
knitr::kable(head(ExampleGDP,3))

Create a first graph

gdp_json <- vt_export_json(vt_input_from_df(ExampleGDP))
vt_d3(gdp_json)

There a couple of settings you can change I

vt_d3(gdp_json,label = FALSE, color_border = "#000000", size_border = "2px", legend = TRUE)

There a couple of settings you can change II

• You can set a seed (in Javascript)!

vt_d3(gdp_json, legend = TRUE, legend_title = "Continents", seed = 1)

A 2nd Example (Canadian Consumer Price Index) I

• Colors can be provided for each cell independently.

data(canada)
canada <- canada[canada$h1=="Canada",]
canada$codes <- canada$h3
canadaH <- vt_export_json(vt_input_from_df(canada,scaleToPerc = FALSE))
vt_d3(canadaH, label=FALSE,width = 400,height = 400)

A 2nd Example (Canadian Consumer Price Index) II

• Colors could be computed according to a numeric variable, e.g. with the scales package.

canada$color <- scales::seq_gradient_pal(low = "#999999",high = "#ffffff")(canada$weight/max(canada$weight))
canadaH <- vt_export_json(vt_input_from_df(canada,scaleToPerc = FALSE))
vt_d3(canadaH, label=FALSE,width = 400,height = 400, color_border = "#000000")

Shiny App I

knitr::include_graphics("shiny1.jpg")

Shiny App II

knitr::include_graphics("shiny2.jpg")

Conclusion

• Package is 90% ready, hopefully available on CRAN soon.
• Great event!! Thanks to the organizers and attendees.
• https://github.com/uros2018/voronoiTreemap

uRosConf/voronoiTreemap documentation built on July 3, 2019, 12:25 a.m.