CODEWITHSUNDEEP
rlistggplot2
user2450223
user2450223

Reputation: 235

GGplot - Inputting a list of data frames

I have a list df_list = (df1,df2,df3) of data frames, all of the same dimension but different column names. I am calling ggplot inside a for loop:

for(i in 1:3){
p = ggplot(df_list[[i]], aes(x=df_list[[i]][1], y=df_list[[i]][3], fill=df_list[[i]][2])) +
  geom_bar(stat='identity', position='dodge')
}

However, R doesn't allow me to do this: I have to actually use the variable names, which means I can't use the loop. Any solutions?

Here is example data:

df1<-data.frame(order=c(1:3,4:6), gender=c('f','f','f','m','m','m'), count= c(10,20,30,20,25,30))
 
 df2<-data.frame(order11=c(2:4,2:4), gender11=c('f','f','f','m','m','m'), count= c(11,21,30,21,25,30))
 
 df3<-data.frame(order22=c(1:3,4:6), gender22=c('f','f','f','m','m','m'), count= c(10,20,31,21,25,30))
 >  df1
  order gender count
1     1      f    10
2     2      f    20
3     3      f    30
4     4      m    20
5     5      m    25
6     6      m    30
>

Upvotes: 1

Views: 767

Answers (2)

Ronak Shah
Ronak Shah

Reputation: 389325

Same logic but a different approach :

library(tidyverse)

df_list = lst(df1,df2,df3) #This is not a typo!!

imap(df_list, ~{
  col <- names(.x)
  ggplot(.x, aes(x=.data[[col[1]]], y=.data[[col[3]]], fill=.data[[col[2]]])) + 
    geom_col(position='dodge') + 
    ggtitle(paste0('Plot for ', .y))
}) -> plot_list

plot_list[[1]]

enter image description here

plot_list[[2]]

enter image description here

Upvotes: 2

Limey
Limey

Reputation: 12585

You may have different column names in your data frames, but the roles of the columns seem to be defined by their index in the data frame. For example, you want the first column in thr data frame to define x values in the plot and so on. This means you can use names() and aes_string() to get what you want.

First define some test data, as you haven't given us any.

library(tidyverse)

df1 <- tibble(x1=rnorm(10), y1=rnorm(10), z1=rep(c("A", "B"), each=5))
df2 <- tibble(x2=rnorm(10), y2=rnorm(10), z2=rep(c("A", "B"), each=5))
df3 <- tibble(x3=rnorm(10), y3=rnorm(10), z3=rep(c("A", "B"), each=5))
dfList <- list(df1, df2, df3)

Now solve the problem

lapply(
  dfList,
  function(df) {
    names <- names(df)
    df %>% ggplot(aes_string(x=names[1], y=names[2], colour=names[3])) + geom_point()
  }
)

Giving

enter image description here enter image description here enter image description here

I've used geom_point() and colour rather than geom_bar() and fill but the same principle will apply.

Upvotes: 5

Related Questions