ggplot2中不同叠加颜色的对比分析

Question

我有一个更大的数据集，类似于以下内容：

example = data.frame(
  person = c(1,1,1, 2,2,2, 3,3,3),
  score =  c(5,1,2, 3,5,1, 1,2,2),
  round =  c(1,2,3, 1,2,3, 1,2,3)
)

我试图绘制它，这样每个人都有他们自己的颜色，并且在堆栈中任何时候都会像渐变一样形成条形。我所能得到的最接近的方法是允许alpha按轮变化：

library(ggplot2)
ggplot(data = example,
       aes(x = reorder(person, score),
           y = score,
           fill = factor(person),
           alpha = as.integer(round))) + 
  geom_col()

如何通过对比改变颜色来创建一个适当的渐变呢？

​

​

Answer 1

你可以直接做

ggplot(data = example, aes(x = reorder(person, score), y = score, 
  fill = interaction(person, as.integer(round)))) + geom_col()

然后用scale_fill_manual定义颜色值来定义渐变。

编辑

简单的例子是：

colornames = sort(levels(interaction(example$person, as.integer(example$round))))
colors = c(
  paste0("red", 1:3),
  paste0("green", 1:3),
  paste0("blue", 1:3)
)

ggplot(data = example, aes(x = reorder(person, score), y = score, 
  fill = interaction(person, as.integer(round)))) + geom_col() +
  scale_fill_manual(values = setNames(colors, colornames))

如果您有很多人员或圆圈，您可能希望使用colorRamp或rgb编程设置颜色。

编辑2

或者使用半透明的白色列来攻击它：

ggplot(data = example, aes(x = reorder(person, score), y = score, fill = factor(person))) + 
  geom_col() + geom_col(aes(alpha = as.integer(round)), fill = "white") +
  scale_alpha(NULL, guide = FALSE, range = c(0, 0.4))

Answer 2

这里的诀窍(受@mikeck的启发)是在原始透明度之下创建坚实的颜色条：

ggplot(data = example, aes(x = reorder(person, score), y = score)) + 
  # Base bar color (black is another good color to try)
  geom_col(fill = "white") +
  # Overlay with the colored transparency
  geom_col(aes(fill = factor(person), alpha = round))