Multiple rows in facet_grid

问题:

I have a dataset that looks roughly like this:

names = tibble(NAME_2=c("Location1","Location2","Location3","Location4"))
dates = tibble(date = seq(as.Date("2015-01-01"), as.Date("2016-12-31"), by="days"))
types = tibble(type = c("comment","post"))
df <- merge(names,dates)
df <- merge(df, types)
zero <- seq(from=0, to=200, by=1)
df$n <- sample(zero, size=nrow(df), replace=TRUE)

Which produces a facet plot like this:

ggplot(data = df, aes(x = date, y = n)) +
  geom_line() +
  facet_grid(type ~ NAME_2, scale = "free_y")

Is it possible to get behavior like ncol=2 in facet_wrap so that Location3 and Location4 appear below Location1 and Location2? In reality I have about 12 locations, which makes it impossible to print on one page and still keep it legible.

回答1:

You could use grid.arrange, as in:

library(gridExtra)
grid.arrange(
  ggplot(data = df[df$NAME_2 %in% c('Location1','Location2'),], aes(x = date, y = n)) +
    geom_line() + xlab(NULL) +
    facet_grid(type ~ NAME_2, scale = "free_y"),
  ggplot(data = df[df$NAME_2 %in% c('Location3','Location4'),], aes(x = date, y = n)) +
    geom_line() +
    facet_grid(type ~ NAME_2, scale = "free_y"),
  nrow=2)

If you're sure that the x axis ranges line up from top to bottom, you could suppress the x axis tick marks/labels on the first plot.