CODEWITHSUNDEEP
rggplot2lapply
Chirayu Chamoli
Chirayu Chamoli

Reputation: 2076

How to access names of the element within lapply function in a plot

I want to generate a boxplot for all the variable in my data frame. But the problem is i cannot get the x axis to be labelled with the variable name which is looping via lapply. Any thoughts.

df=data.frame(a=rnorm(100), b=rnorm(100), c=rnorm(100), d=rep(1:2,50)) # data
df$d=as.factor(df$d)

Function for box plot

Boxplotfun=function(x,i){
  if(is.numeric(x)) {
    p <- ggplot(df, aes(d, x, group=d)) + geom_boxplot(outlier.colour = "red", fill = c("gold","darkgreen")) + coord_flip() + theme_light() + ggtitle(paste(names(x)))
    plot(p)
  } 
}

l=lapply(df[-ncol(df)], Boxplotfun)

Upvotes: 1

Views: 187

Answers (2)

Choubi
Choubi

Reputation: 680

Instead of using the dataframe as first argument to lapply, you can use the names of the dataframe directly and just refer to your dataframe inside of your plot function.

Instead of using aes, I have also used aes_string so that I can pass "x" inside and ggplot understands that I am not looking for variable x but rather for the variable whose name "x" refers to

Boxplotfun=function(x) {
  if(is.numeric(df[, x])) {
    p <- ggplot(df, aes_string("d", x, group="d")) + 
      geom_boxplot(outlier.colour = "red", fill = c("gold","darkgreen")) + 
      coord_flip() + theme_light() + 
      ggtitle(x)
    plot(p)
  } 
}

ll <- lapply(names(df[-ncol(df)]), Boxplotfun)

Edit: If you want the legend to appear, you should, as Axeman's answer has shown, specify fill inside your aes and use scale_fill_manual to pick the colors:

Boxplotfun=function(x) {
  if(is.numeric(df[, x])) {
    p <- ggplot(df, aes_string("d", x, fill="d")) + 
      geom_boxplot(outlier.colour = "red")+
      scale_fill_manual(values = c("gold","darkgreen")) +
      coord_flip() + theme_light() + 
      ggtitle(x)
    plot(p)
  } 
}

Upvotes: 1

Axeman
Axeman

Reputation: 35242

Just to make Richard's comment into an answer:

df <- data.frame(a=rnorm(100), b=rnorm(100), c=rnorm(100), d=rep(1:2,50)) # data
df$d <- as.factor(df$d)

df2 <- tidyr::gather(df, 'var', 'val', -d)
ggplot(df2, aes(d, val, fill = d)) + 
  geom_boxplot(outlier.colour = "red") +
  scale_fill_manual(values = c("gold","darkgreen")) +
  coord_flip() + 
  theme_light() +
  facet_grid(var~.)

enter image description here

Upvotes: 2

Related Questions