Iterate over a categorical and a numerical data frames and ploting a bar chart in r-CodePudding

I have a numeric data frame call A and a categoriacal one call B For every two variables (one from B and one from A) I want to plot a bar chart.

My try:

y <- unlist(lapply(D, Negate(is.numeric)))
x <- unlist(lapply(D, is.numeric))    
A <- D[,x]
B <- D[,y]
A <- as.data.frame(A)
B <- as.data.frame(B)

for(i in (1:ncol(A)))
{
  for(j in (1:ncol(B)))
  {
    ggplot(D, aes(x = A[,i], y = B[,j]))  
    geom_bar(fill='red') 
  }
}

But is not plotting anything. I'm not sure if it's necessary to save each plot in a vector and then plot them with another for. Any suggestions would be great!

CodePudding user response：

You will need to wrap the ggplot statement inside the print() function to generate output while inside a loop:

for(i in (1:ncol(A)))
{
  for(j in (1:ncol(B)))
  {
    g<- ggplot(D, aes(x = A[,i], y = B[,j]))  
             geom_bar(fill='red') 
    print(g)
  }
}

CodePudding user response：

I think you need to replace geom_bar() with geom_col() or to add stat = identity inside geom_bar().

You can also avoid using the for loop by using pmap from package purrr:

 df <- expand.grid(
    i = 1:ncol(A),
    j = 1:ncol(B)
)

purrr::pmap(
    list(df$i, df$j),
    function(.i, .j){
        tmp <-ggplot(D, aes(x = A[,.i], y = B[,.j]))  
            geom_col(fill='red')
        print(tmp)
    }
)