Forum Discussion

ross27's avatar
ross27
Advocate II
9 years ago
Solved

R Visual with transparent background

Is it possible to create an R-Visual with a transparent background i.e. so that you can see an image added to the page background behing the plot the R-Visual generates?   Specifically I'm using gg...
  • arvidb's avatar
    arvidb
    7 years ago

    I don't know whether this will work with ggplot2 but in a visual I made based on the fmsb package I was able to work around this issue by using the par() function before the plotting was called:

     

    # generic par function for graphical elements, call before the plotting
    par(bg="transparent")

    # this is wear the plotting of the visual actually happens
radarchart(data)

     

v-haibl-msft's avatar
v-haibl-msft
Microsoft Employee
9 years ago

ross27

 

There is no option in Format pane to make the R visual background to be transparent. Only small part (around the R visual) can be transparent as below.

 

But we should be able to set bg="transparent" to make it to be transparent like below. Could you please share your R scripts written in the R script editor? Please also take a look at this link to see if it helps.

 

Best Regards,

Herbert

  • ross27's avatar
    ross27
    Advocate II
    9 years ago

    Thanks Herbert.  This is the final part of my script:

     

    g <- g + theme_void() +
theme(
    axis.text.x = element_text(angle = 45,size=11,vjust=0.9),
    legend.position="none",
    plot.background = element_rect(fill="transparent", color=NA),
    panel.background = element_rect(fill="transparent", color=NA)
  )

par(bg="transparent")

g

    But as you can see I still have a white background:

     

     

    • v-haibl-msft's avatar
      v-haibl-msft
      Microsoft Employee
      9 years ago

      ross27

       

      How about the result if update the script as below? If problem still persists, could you please share your .pbix file?

       

      panel.background = element_rect(fill = "transparent",colour = NA), 
panel.grid.minor = element_blank(), 
panel.grid.major = element_blank(),
plot.background = element_rect(fill = "transparent",colour = NA))
      panel.background = theme_rect(fill = "transparent",colour = NA),
panel.grid.minor = theme_blank(), 
panel.grid.major = theme_blank(),
plot.background = theme_rect(fill = "transparent",colour = NA)

      Best Regards,

      Herbert

      • ross27's avatar
        ross27
        Advocate II
        9 years ago

        I still have the same issue.  I can't see how to upload a pbix, but here's the full code behind the visual.

         

        "audience" is a text column providing the labels for my x-axis

        "overlap" is an integer used to size the data points

        "gindex" is a decimal, plotted on the y-axis

        "indexColor" is either 1 or -1 and is used to colour the circles.

         

        dataset <- data.frame(audience, overlap, gindex, indexColor)

library(ggplot2)

rescale <- function(x,over=dataset$overlap,minSize=5,maxSize=40) {
  x <- log(x)
  over <- log(over)
  
  rng <- range(over)
  
  ((x-rng[1])/(rng[2]-rng[1])) * (maxSize-minSize) + minSize
  
}

g <- ggplot(dataset,aes(audience,gindex))

#Lines

g <- g + geom_segment(aes(xend=audience,yend=0),linetype="dashed",size=1.5)

#Circles

g <- g + geom_point(aes(size=rescale(overlap),colour=indexColor)) +
  scale_size_identity() +
  scale_colour_gradientn(colours=c("#FF0000","#00FF00"))

#Ensure circles fit at top of plot (20 = maxSize used in rescale()/2) and
# remove gap between y-axis bottom of plot area

g <- g + 
    scale_y_continuous(limits = c(0,max(dataset$gindex)+20), expand = c(0, 0)) 

#Labels for index values
g <- g + geom_text(aes(label=gindex), fontface = "bold")

#Remove unwanted styling,
#   Adjust axis labels
#
g <- g + theme_void() +
theme(
    axis.text.x = element_text(angle = 45,size=11,vjust=0.9),
    legend.position="none",
    panel.background = element_rect(fill = "transparent",colour = NA),
    panel.grid.minor = element_blank(), 
    panel.grid.major = element_blank(),
    plot.background = element_rect(fill = "transparent",colour = NA)
  )

par(bg="transparent")

g