Plotting multiple plots using ggplot and facet_wrap

Question

Sample data

set.seed(123)
df <- data.frame(loc.id = rep(c(1:3), each = 4*10), 
                        year = rep(rep(c(1980:1983), each = 10), times = 3),
                        day = rep(1:10, times = 3*4),
                        x = sample(123:200, 4*3*10, replace = T),
                        start = 123,
                        end = 200)

I want to save the plot of each loc.id for all years in a single page using facet_wrap and each loc.id in separate pages as a pdf. Following loop does this:

loc.vec <- 1:3

pdf("my.pdf")

for(l in  seq_along(loc.vec)){

  loc.id <- loc.vec[l]
  df.sub <- df[df$loc.id == loc.id,]

  pp <- ggplot(df.sub,aes(x = day, y = x)) + geom_line() +
          facet_wrap(~year) + 
          geom_vline(aes(xintercept = df.sub$start)) + 
          geom_vline(aes(xintercept = df.sub$end))
  print(pp)
}
dev.off()

Can I achieve without the loop?

Thanks

The no. of loc.id I have is way too many. So was wondering if there is any quicker solution to this. Don't mind using a loop — 89_Simple
– 89_Simple, Commented May 31, 2018 at 15:16
It wasn't my downvote, but I can hazard a guess why. Your question includes code that works fine. There's no way around needing some kind of loop here, and your original choice of for is as good as any from a functional viewpoint. You may get a range of answers suggesting to use map or apply or by or any of the many other ways of looping in R. But none of them changes the underying fact that you need to repeat an operation several times iterating over loc.id. In other words you need a loop and you already have one that works just fine. — dww
– dww, Commented May 31, 2018 at 20:13
Fair point. I guess I was more talking about from a speed perspective i.e. if I had 1 million loc.id, would this loop will still be faster compared to the solution below. — 89_Simple
– 89_Simple, Commented Jun 1, 2018 at 9:39

bVa · Accepted Answer · 2018-05-31 15:19:44Z

1

Here is a solution using purrr:

library(tidyverse)

f_plot <- function(id) {
  df %>%
    filter(loc.id == id) %>%
    ggplot(., aes(x = day, y = x)) + 
    geom_line() +
    facet_wrap(~year) + 
    geom_vline(aes(xintercept = start)) + 
    geom_vline(aes(xintercept = end))
}

pdf("my2.pdf")
map(loc.vec, f_plot)
dev.off()

answered May 31, 2018 at 15:19

bVa

3,9681 gold badge16 silver badges22 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

Parfait · Accepted Answer · 2018-05-31 15:46:14Z

1

Consider by (being the object-oriented wrapper to tapply) to slice dataframe by the loc.vec factor and run subsets through plot:

process_plots <- function(df.sub) {    
  ggplot(df.sub, aes(x = day, y = x)) + 
    geom_line() + facet_wrap(~year) + 
    geom_vline(aes(xintercept = df.sub$start)) + 
    geom_vline(aes(xintercept = df.sub$end))      
}

pdf("my.pdf")    
by(df, df$loc.vec, process_plots)
dev.off()

answered May 31, 2018 at 15:46

Parfait

108k19 gold badges103 silver badges138 bronze badges

Collectives™ on Stack Overflow

Plotting multiple plots using ggplot and facet_wrap

2 Answers 2

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related