CODEWITHSUNDEEP
rggplot2
Nancy
Nancy

Reputation: 4109

Suppress message from geom_line with only one point

I'm iterating through multiple data sets to produce line plots for each set. How can I prevent ggplot from complaining when I use geom_line over one point?

Take, for example, the following data:

mydata = data.frame(
  x = c(1, 2),
  y = c(2, 2),
  group = as.factor(c("foo", "foo"))
)

Creating line graph looks and works just fine because there are two points in the line:

ggplot(mydata, aes(x = x, y = y)) + 
  geom_point() + 
  geom_line(aes(group = group))

However, plotting only the first row gives the following message:

geom_path: Each group consists of only one observation. Do you need to adjust the group aesthetic?

ggplot(mydata[1,], aes(x = x, y = y)) + 
      geom_point() + 
      geom_line(aes(group = group))

Some of my figures will only have a single point and the messages cause hangups in the greater script that produces these figures. I know the plots still work, so my concern is avoiding the message. I'd also like to avoid using suppressWarnings() if possible in case another legitimate and unexpected issue arises.

Upvotes: 4

Views: 1065

Answers (4)

E. Nygaard
E. Nygaard

Reputation: 181

I found @vinnief's answer helpful when experiencing a similar problem with multiple annoying "geom_line(): Each group consists of only one observation." messages.

Posting my solution illustrating how the original poster could implement the strategy described by John Mackintosh on community.RStudio.com, in case someone else finds it useful:

mydata <- data.frame(
  x = c(1, 2),
  y = c(2, 2),
  group = as.factor(c("foo", "foo"))
)

# No annoying warning:
ggplot(mydata, aes(x = x, y = y)) + 
  geom_point() + 
  geom_line(aes(group = group))

# Plotting only the first row results in warning:
ggplot(mydata[1,], aes(x = x, y = y)) +
geom_point() + 
  geom_line(aes(group = group))

geom_point() does not complain if there is only one observation, but geom_line() does - it expects to find at least two points between which to draw a line.

So we will firstly perform geom_point(), followed by geom_line(), but with a twist. We will supply a filtered version of mydata to geom_line(). By removing rows from mydata where y only occurs once in the entire data frame, we ensure that the data supplied to geom_line() does not trigger the warning message, while at the same time retaining any rows with x and y coordinates that actually form lines that need to be plotted.

ggplot(mydata[1,], aes(x = x, y = y)) + 
  geom_point() + 
  geom_line( 
    # Filtered data frame:
    data = mydata[1,] %>% dplyr::add_count(y) %>% filter(n > 1)
    # # It works with the complete data frame as well:
    # data = mydata %>% dplyr::add_count(y) %>% filter(n > 1)
    , aes(group = group)
  )

Using a data frame with zero rows does not appear to upset geom_line().

Upvotes: 1

vinnief
vinnief

Reputation: 821

On the community.RStudio.com, John Mackintosh suggests a solution which worked for me:

Freely quoting:

Rather than suppress warnings, change the plot layers slightly.

  1. Facet wrap to create empty plot

  2. Add geom_point for entire data frame

  3. Subset the dataframe by creating a vector of groups with more than one data point, and filtering the original data for those groups. Only plot lines for this subset.

Details and example code in the followup of the link above.

Upvotes: 1

John J.
John J.

Reputation: 1748

Per an answer to this question: suppressMessages(ggplot()) fails because you have to wrap it around a print() call of the ggplot object--not the ggplot object itself. This is because the warning/message only occurs when the object is drawn.

So, to view your plot without a warning message run:

p <- ggplot(mydata[1,], aes(x = x, y = y)) + 
  geom_point() + 
  geom_line(aes(group = group))

suppressMessages(print(p))

Upvotes: 4

Lars Arne Jordanger
Lars Arne Jordanger

Reputation: 671

I think the following if-else solution should resolve the problem:

if (nrow(mydata) > 1) {
    ggplot(mydata, aes(x = x, y = y)) + 
        geom_point() + 
        geom_line(aes(group = group))
} else {
    ggplot(mydata, aes(x = x, y = y)) + 
        geom_point()
}

Upvotes: 0

Related Questions