如何在图上反转y轴

I would like to know how to make a plot in R where the y-axis is inverted such that the plotted data appears in what would be the fourth quadrant (IV) of a cartesian plane, as opposed to the first (I) quadrant.

我想知道如何在R中绘制一个图，其中y轴被反转，使得绘制的数据出现在笛卡尔平面的第四象限（IV）中，而不是第一个（I）象限。

For reference, the plot I am trying to make looks very similar to the following (source):

作为参考，我试图制作的图看起来非常类似于以下（源）：

I have found a number of questions online pertaining to reversing the numbering on the y-axis, but these all still plot the data in the first quadrant. Can anyone suggest how I might produce a plot similar to that shown above?

我在网上发现了许多关于反转y轴编号的问题，但这些问题仍然在第一象限中绘制数据。任何人都可以建议我如何制作类似于上面所示的情节？

3 个解决方案

#1

Just to provide a worked out answer, following the comments of @timriffe and @joran...

根据@timriffe和@joran的评论，提供一个得出的答案......

Use the function for minor log ticks from this answer:

使用此答案的次要日志刻度函数：

minor.ticks.axis <- function(ax,n,t.ratio=0.5,mn,mx,...){

  lims <- par("usr")
  if(ax %in%c(1,3)) lims <- lims[1:2] else lims[3:4]

  major.ticks <- pretty(lims,n=5)
  if(missing(mn)) mn <- min(major.ticks)
  if(missing(mx)) mx <- max(major.ticks)

  major.ticks <- major.ticks[major.ticks >= mn & major.ticks <= mx]

  labels <- sapply(major.ticks,function(i)
    as.expression(bquote(10^ .(i)))
                   )
  axis(ax,at=major.ticks,labels=labels,...)

  n <- n+2
  minors <- log10(pretty(10^major.ticks[1:2],n))-major.ticks[1]
  minors <- minors[-c(1,n)]

  minor.ticks = c(outer(minors,major.ticks,`+`))
  minor.ticks <- minor.ticks[minor.ticks > mn & minor.ticks < mx]


  axis(ax,at=minor.ticks,tcl=par("tcl")*t.ratio,labels=FALSE)

}

Make some reproducible example data:

制作一些可重现的示例数据：

x <- 1:8
y <- 10^(sort(runif(8, 1, 10), decreasing = TRUE))

Plot without axes:

没有轴的情节：

plot(x, log10(y), # function to plot
xlab="",          # suppress x labels
type = 'l',       # specify line graph
xlim = c(min(x), (max(x)*1.3)),  # extend axis limits to give space for text annotation
ylim = c(0, max(log10(y))),      # ditto
axes = FALSE)    # suppress both axes

Add fancy log axis and turn tick labels right way up (thanks @joran!):

添加花式日志轴并向上翻转刻度标签（感谢@joran！）：

minor.ticks.axis(2, 9, mn=0, mx=10, las=1)

Add x-axis up the top:

在顶部添加x轴：

axis(3)

Add x-axis label (thanks for the tip, @WojciechSobala)

添加x轴标签（感谢提示，@ WojciechSobala）

mtext("x", side = 3, line = 2)

And add an annotation to the end of the line

并在行尾添加注释

text(max(x), min(log10(y)), "Example", pos = 1)

Here's the result:

这是结果：

#2

Answering the question in the title, the best/easiest way to invert the axis is to flip the limit variables around:

回答标题中的问题，反转轴的最佳/最简单方法是翻转限制变量：

> plot(1:10, xlim=c(1,10));

> plot(1:10, xlim=c(10,1));

> plot(1:10, ylim=c(10,1));

Doing it this way means that you don't need to mess around with axes that are different from the image coordinates.

这样做意味着您不需要使用与图像坐标不同的轴。

This can be combined with the 'xaxt="n"' parameter and an additional axis command to place an axis on another side:

这可以与'xaxt =“n”'参数和另一个轴命令结合使用，将轴放在另一侧：

> plot(1:10, ylim=c(10,1), xaxt="n"); axis(3);

#3

It's now quite easy to reverse the y-axis using scale_y_reverse and specify position = "top" for the x-axis in ggplot2

现在很容易使用scale_y_reverse来反转y轴，并在ggplot2中为x轴指定position =“top”

Example

例

library(ggplot2)
library(scales)

set.seed(99)
Date <- seq(from = as.Date("2017-12-01"), to = as.Date("2017-12-15"),
             by = "days")
Flux <- runif(length(Date), 1, 10000)
Flux_df <- data.frame(Date, Flux)

p1 <- ggplot(Flux_df, aes(Date, Flux)) +
  geom_col() +
  xlab("") +
  scale_x_date(position = "top", breaks = pretty_breaks(), expand = c(0, 0)) +
  scale_y_reverse(expand = expand_scale(mult = c(0.2, 0))) +
  theme_bw(base_size = 16) +
  theme(panel.border = element_blank(),
        panel.grid.major.x = element_blank(),
        panel.grid.minor = element_blank(),
        axis.line = element_line()) +
  theme(legend.position = "none")
p1

If we want both logarithmic and reverse axis, we need a workaround suggested here as ggplot2 does not have that option atm

如果我们想要对数轴和反向轴，我们需要在此处建议的解决方法，因为ggplot2没有该选项atm

reverselog_trans <- function(base = exp(1)) { 
  trans <- function(x) -log(x, base) 
  inv <- function(x) base^(-x) 
  scales::trans_new(paste0("reverselog-", format(base)), trans, inv, 
                    scales::log_breaks(base = base), domain = c(1e-100, Inf)) 
}


p1 + scale_y_continuous(trans = reverselog_trans(10),
                        breaks = scales::trans_breaks("log10", function(x) 10^x),
                        labels = scales::trans_format("log10", scales::math_format(10^.x)),
                        expand = expand_scale(mult = c(0.2, 0))) +
  annotation_logticks()