How can I combine group and stack bar chart with layout.shape in background using R Plotly?

Question

I am trying to create a visualization where I need use both stack and group bar charts. The data I am using has thousand of rows, the below one is the subset of that.

 Month      Week   Cat   n
_____________________________
 2019-Dec    4      A    17
 2019-Dec    4      B    6
 2019-Dec    5      A    21
 2019-Dec    5      C    10
 2020-Jan    1      A    19
 2020-Jan    1      B    20
 2020-Jan    1      C    12

The plot I want is something like below:

Where it's group by Month, stack by Cat and to distinguish different Week I want to add rectangular shapes in the background (https://plotly.com/r/shapes/) for different weeks.

Thanks in advance!

This is currently not implemented. You can track this [GitHub issue](https://github.com/plotly/plotly.js/issues/4914) for progress. — ismirsehregal, Oct 30 '20 at 08:51
Thanks for the information. Is it anyway possible using ggplot? As it is always possible to transform ggplot to plotly. — Williams86, Oct 30 '20 at 09:08

score 2 · Accepted Answer · answered Oct 30 '20 at 09:25

As mentioned in the comments stacked+grouped bar charts aren't implemented yet in plotly. Here is a ggplot2 / ggplotly workaround:

library(plotly)
library(ggplot2)

DF <- data.frame(
  stringsAsFactors = FALSE,
             Month = c("2019-Dec","2019-Dec",
                       "2019-Dec","2019-Dec","2020-Jan","2020-Jan","2020-Jan"),
              Week = c(4L, 4L, 5L, 5L, 1L, 1L, 1L),
               Cat = c("A", "B", "A", "C", "A", "B", "C"),
                 n = c(17L, 6L, 21L, 10L, 19L, 20L, 12L)
)

p <- ggplot(DF, aes(x = Week, y = n, fill = Cat)) + geom_bar(stat = 'identity', position = 'stack') + facet_grid( ~ Month) + theme_bw()

ggplotly(p)

How can I combine group and stack bar chart with layout.shape in background using R Plotly?

1 Answers1