CODEWITHSUNDEEP
rfunctionr-markdown
user14176250
user14176250

Reputation: 55

summary table with chart in output

this is my desired output , my desired output is to get summary table with chart in single output enter image description here


  T1 <- as.data.frame(table(data[[var]]))
  bar_chart <- barchart(data[[var]]))

Upvotes: 0

Views: 425

Answers (1)

LHI
LHI

Reputation: 36

enter image description here

I can't seems to reproduce your example, but I understand you want to combine a table and a plot. I used some hypothetical data as example.

You can create the table as grob object using tableGrob function from gridExtra package, convert the ggplot to a grob object using ggplotGrob from ggplot2 package, and then combine them in different ways.

  1. annotation_custom function from ggplot2 package: Here you can feck an empty ggplot and arrange the table and the plot in different regions of the empty plot (see code example below)
# making the plot 
plot_df <- data.frame(x=LETTERS[1:5], y=1:5)
plot <- ggplot(plot_df, aes(x, y))+
  geom_col()

# making the table
table_df <- data.frame(x=LETTERS[1:10], y=1:10)
tabl <- tableGrob(d = table_df)

# making the feck plot that will contain the plot and the table
feck_plot_df <- data.frame(x=0:1, y=0:1)
p <- ggplot(feck_plot_df)+
  geom_blank()+
  annotation_custom(grob = tabl, 
                    xmin = -Inf, xmax = 0.1, ymin = 0.15, ymax = 0.9)+
  annotation_custom(grob = ggplotGrob(plot), 
                    xmin = 0.1, xmax = 1, ymin = -Inf, ymax = 1)

# saving the combined plot 
ggsave(plot = p, filename = 'combined_plot.png', device = 'png', dpi = 300,
       width = 7.04, height = 3.77)

You can also coerce the flexible table to raster flexible::as_raster, then to grob using grid::rasterGrob and arrange things in an empty ggplot. Based on your updates, below is an attempt to arrange the 3 plots.

tabl <- tab_std(data=df, var = "col1", Name_of_variable = "state")
chrt <- donut(df,"col1")

tabl <- tab_std(data=df, var = "col1", Name_of_variable = "state")
bplt <- single_bar(df,"col1", SORT = TRUE)

ggplot() + 
  theme_void()+
  annotation_custom(grid::rasterGrob(as_raster(tabl)), xmin=-Inf, xmax=0.5, ymin=0.5, ymax=Inf)+
  annotation_custom(ggplotGrob(chrt), xmin=-Inf, xmax=0.5, ymin=-Inf, ymax=0.5)+
  annotation_custom(ggplotGrob(bplt), xmin=0.5, xmax=Inf, ymin=-Inf, ymax=Inf)
  1. Grid-based layouts functions from gridExtra package: Here, you may want to try grid.arrange, arrangeGrob, or gtable_cbind.
tabl <- tab_std(data=df, var = "col1", Name_of_variable = "state")
chrt <- donut(df,"col1")

tabl <- tab_std(data=df, var = "col1", Name_of_variable = "state")
bplt <- single_bar(df,"col1", SORT = TRUE)


grid.arrange(grid::rasterGrob(as_raster(tabl)), ggplotGrob(chrt), ggplotGrob(bplt),
             layout_matrix = cbind(rbind(1,2), rbind(3,3)))

Once you get the plot saved somewhere, then you can just include it into your rmarkdown using include_graphics function from knitr package. You can also directly include it without saving.

Upvotes: 2

Related Questions