How do I write a function to plot a line graph for each factor in a dataframe?

Question

I have a dataframe, the head of which looks like this:

|trackName            | week| sum|
|:--------------------|----:|---:|
|New Slang            |    1| 493|
|You're Somebody Else |    1| 300|
|Mushaboom            |    1| 297|
|San Luis             |    1| 296|

I am interested in plotting a line graph for each of the 346 unique trackNames in the dataframe, with week on the x-axis and sum on the y-axis. To automate this process, I wrote the following function:

charts <- function(df) {
  songs <- df
  lim <- nrow(songs)
  x <- 1
  song_names <- as_tibble(unique(songs$trackName))
  while (x <= lim) {
    song <- song_names[x, 1]
    plot.name <- paste(paste(song), "plot.png", sep = "_")
    songs %>% filter(trackName == paste(song[x, 1])) %>%
      ggplot(., aes(x = week, y = sum), group = 1) +
      geom_line() +
      labs(
        x = "Week",
        y = "Sum of Listens",
        title = paste("Week by Week Listening Interest for", song, sep = " "),
        subtitle = "Calculated by plotting the sum of percentages of the song listened per week, starting from first listen"
      ) +
      ggsave(plot.name,
             width = 20,
             height = 15,
             units = "cm")
    x <- x + 1
  }
}

However when I run charts(df), only the following error shows up and then it quits:

> charts(mini)
geom_path: Each group consists of only one observation. Do you need to
adjust the group aesthetic?
>

What am I doing wrong here and what does this error mean?

A sample of the dataframe in DPUT format:

structure(list(trackName = c("New Slang", "You're Somebody Else", 
"Mushaboom", "San Luis", "The Trapeze Swinger", "Flightless Bird, American Mouth", 
"tere bina - Acoustic", "Only for a Moment", "Upward Over the Mountain", 
"Virginia May", "Never to Be Forgotten Kinda Year", "Little Talks", 
"Jhak Maar Ke", "Big Rock Candy Mountain", "Sofia", "Aaoge Tum Kabhi", 
"Deathcab", "Dil Mere", "Choke", "Phir Le Aya Dil", "Lucille", 
"tere bina - Acoustic", "Dil Mere", "Only for a Moment", "This Is The Life", 
"San Luis", "Main Bola Hey!", "Choo Lo", "Yeh Zindagi Hai", "Aaftaab", 
"Never to Be Forgotten Kinda Year", "Khudi", "Flightless Bird, American Mouth", 
"Mere Bina", "Simple Song", "Dil Haare", "Dil Hi Toh Hai", "You're Somebody Else", 
"Sofia", "Who's Laughing Now", "Main Bola Hey!", "Lucille", "Eenie Meenie", 
"tere bina - Acoustic", "New Slang", "Aaftaab", "Mamma Mia", 
"July", "Yeh Zindagi Hai", "Someone You Loved"), week = c(1, 
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 
2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 3, 3, 3, 
3, 3, 3, 3, 3, 3, 3), sum = c(493, 300, 297, 296, 292, 234, 214, 
200, 200, 197, 192, 187, 185, 181, 175, 172, 141, 119, 106, 103, 
579, 574, 501, 462, 428, 378, 320, 307, 306, 301, 301, 300, 300, 
300, 300, 300, 296, 294, 251, 242, 3534, 724, 696, 512, 479, 
400, 302, 300, 300, 300)), row.names = c(NA, -50L), class = c("tbl_df", 
"tbl", "data.frame"))

What were you intending with group = 1? And can you explain what you mean or intend to produce by "a line graph"? — IRTFM
– IRTFM, Commented May 14, 2021 at 3:40
@IRTFM It is based on this answer, which discusses the same error. I would just like to be able to plot the week and sum using the geom_line() function for each of the trackNames in the dataframe. — Aman
– Aman, Commented May 14, 2021 at 3:46
@IanCampbell Aye no such intention! I realize it was bad practice to include more dependencies than required. — Aman
– Aman, Commented May 14, 2021 at 4:02

Ian Campbell · Accepted Answer · 2021-05-14 03:58:23Z

1

How about using purrr::walk instead?

library(tidyverse)
library(hrbrthemes)
walk(unique(songs$trackName),
     ~{ggsave(plot = ggplot(filter(songs, trackName == .x), aes(x = week, y = sum), group = 1) +
                       geom_line(color = ft_cols$yellow) +
                       labs(x = "Week", y = "Sum of Listens", title = paste("Week by Week Listening Interest for", .x, sep = " "),
                       subtitle = "Calculated by plotting the sum of percentages of the song listened per week, starting from first listen") +
                       theme_ft_rc(),
             file = paste0(.x,"_plot.png"), width = 20, height = 15, units = "cm")})

Note: the question was subsequently edited to remove the hrbrthemes package requirement.

edited May 14, 2021 at 3:58

answered May 14, 2021 at 3:53

Ian Campbell

25.2k14 gold badges43 silver badges62 bronze badges

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

1 Comment

Aman Over a year ago

This works fantastically! I definitely need to learn the purrr package better. Thank you so much.

Ronak Shah · Accepted Answer · 2021-05-14 03:59:37Z

1

You can split the dataset for each trackName and create a png file for it.

library(tidyverse)

charts <- function(df) {
  df %>%
    group_split(trackName) %>%
    map(~{
      track <- first(.x$trackName)
      ggplot(.x, aes(x = factor(week), y = sum, group = 1)) +
        geom_line() +
        labs(
          x = "Week",
          y = "Sum of Listens",
          title = paste("Week by Week Listening Interest for", track),
          subtitle = "Calculated by plotting the sum of percentages of the song listened per week, starting from first listen"
        ) -> plt
      
      ggsave(paste0(track,'.png'), plt, width = 20, height = 15, units = "cm")
    })
}

charts(songs)

answered May 14, 2021 at 3:59

Ronak Shah

391k20 gold badges173 silver badges237 bronze badges

1 Comment

Aman Over a year ago

Thank you for your answer! This makes sense, I was not aware of the group_split() function. I shall look into this and hopefully use it the next time :)

Collectives™ on Stack Overflow

How do I write a function to plot a line graph for each factor in a dataframe?

2 Answers 2

1 Comment

1 Comment

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

1 Comment

1 Comment

Your Answer

Sign up or log in

Post as a guest

Linked

Related