如何用facet_grid()创建Y轴等距且美观的自由刻度断点

Hi Malte, I get it—getting those 3 clean, evenly spaced y-axis breaks per facet can be tricky when using scales = "free_y". Let's tweak your code to fix that!

The Problem with Your Current Function #

Your equal_breaks() function adds a margin via the s parameter and rounds to 2 decimal places, which can lead to unappealing, uneven-looking breaks. Instead, we can leverage R's built-in pretty() function to generate "nice" (human-readable) break points while ensuring exactly 3 evenly spaced values per facet.

Improved Break Function #

Here's a revised function that prioritizes both uniformity and readability:

nice_3_breaks <- function() {
  function(x) {
    # Get initial pretty breaks for the facet's y-range
    pretty_breaks <- pretty(x, n = 3)
    
    # Handle cases where pretty returns more than 3 breaks (pick first, middle, last)
    if (length(pretty_breaks) > 3) {
      mid_idx <- floor(length(pretty_breaks) / 2)
      pretty_breaks <- pretty_breaks[c(1, mid_idx, length(pretty_breaks))]
    }
    # Handle edge cases where pretty returns fewer than 3 (force 3 evenly spaced)
    else if (length(pretty_breaks) < 3) {
      pretty_breaks <- seq(min(x), max(x), length.out = 3)
      # Round to 1 decimal place for cleanliness (adjust as needed)
      pretty_breaks <- round(pretty_breaks, 1)
    }
    
    pretty_breaks
  }
}

Updated Plot Code #

Swap out your old breaks function with this new one in your ggplot code:

# Packages
library(dplyr)
library(ggplot2)
library(scales)

# Test Data
set.seed(123)
result_df <- data.frame(
  variable = rep(c(1,2,3,4), each = 4),
  mode = rep(c(1,2), each = 2),
  treat = rep(c(1,2)) %>% as.factor(),
  mean = rnorm(16, mean = .7, sd = 0.2),
  x = abs(rnorm(16, mean = 0, sd = 0.5))
) %>% mutate(lower = mean - x,upper = mean + x)

## Plot with Fixed Breaks
result_df %>% 
  ggplot(aes(y = mean*100, x = treat)) + 
  geom_pointrange(aes(ymin = lower*100, ymax = upper*100), shape = 20) + 
  facet_grid(variable ~ mode, scales = "free_y")+ 
  scale_y_continuous(breaks = nice_3_breaks())+ 
  labs(x = "", y = "")

How This Works #

• The pretty() function automatically selects break points that are easy to read (like multiples of 10, 5, etc.) instead of arbitrary decimals.
• We adjust for cases where pretty() returns more or fewer than 3 breaks to ensure exactly 3 evenly spaced values per facet.
• The rounding step keeps the breaks clean without sacrificing uniformity.

This should give you the clean, consistent y-axis breaks you're looking for across all your facets!

内容的提问来源于stack exchange，提问作者m_ky