Log10顶部和Y轴geom_line上的x轴

本文介绍了Log10顶部和Y轴geom_line上的x轴的处理方法，对大家解决问题具有一定的参考价值，需要的朋友们下面随着小编来一起学习吧！ 问题描述 depth = c（1.6,2.6,3.6,4.6）我已经在ggplot2中创建了一个图形，如下所示： （5.6,6.6,7.6,8.6） ri df = data.frame（depth，ri） library（ggplot2） m m + scale_x_log10（Richardson Number，breaks = c（0.1,0.25,0.5,1,5,10,50））+ scale_y_reverse（深度（m）） 这是输出： 我正在尝试做的事情是沿着顶部有x轴，并且还包括一个geom_line，就像我添加的那个这里（手动绘制）。 据我所知，很难让ggplot2移动坐标轴，并且我试图在ggvis中重现此图，但我是你才能得到我需要的log10量表。有没有什么办法可以用R来创建我所瞄准的图表？ 解决方案 更新解决方案 从 ggplot 2.2.0，可以在面板的顶部（和/或在面板的右侧） ） library（ggplot2） depth = c（1.6,2.6,3.6,4.6,5.6 （6.6,7.6,8.6） ri df = data.frame（depth，ri） m scale_x_log10（Richardson Number，breaks = c（0.1,0.25,0.5,1,5,10,50）， $ = b 原始解决方案原始的，经过一点更新到ggplot version2.2.0。 使用 gtable 函数可以移动轴。根据@ Walter的回答调整代码，基本思想是：获取轴（轴文本和刻度标记）;反转轴文本和刻度标记;在绘图面板正上方的gtable布局中添加一行;将修改后的轴插入新行。 library（ggplot2） library（gtable） library （电网） 深度= c（1.6,2.6,3.6,4.6,5.6,6.6,7.6,8.6） ri df = data.frame（depth，ri） m scale_x_log10（Richardson Number，breaks = c（0.1,0.25,0.5,1,5,10,50））+ scale_y_reverse（Depth（m））+ geom_path（） ＃获取ggplot grob g1 ##获取图形面板在g1 中的位置pp ＃标题grobs有边距。 ＃边距需要交换。 ＃交换边界的函数 - 从cowplot包取得的＃：＃https://github.com/wilkelab/cowplot/blob/master/R/switch_axis.R vinvert_title_grob< - function（grob）{ heights< - grob $ heights grob $ heights [1]< - heights [3] grob $ heights [3]< - 高度[1] grob $ vp [[1]] $ layout $ heights [1]< - 高度[3] grob $ vp [[1]] $ layout $ heights [3] < - 高度[1] grob $ children [[1]] $ hjust< - 1 - grob $ children [[1]] $ hjust grob $ children [[1 ]] $ vjust< - 1 - grob $ children [[1]] $ vjust grob $ children [[1]] $ y grob } ＃获取xlab和swap边距索引< - 其中（g1 $ layout $ name ==xlab-b ） xlab xlab ＃将xlab放在g1 的顶部g1 g1 ＃获取x轴（轴线，刻度线和刻度标记）索引 xaxis ＃交换轴刻度和刻度标记刻度< - xaxis $ children [[2]] ticks $ heights< - rev（ticks $ heights） ticks $ grobs< - rev（ticks $ grobs） ＃移动勾号＃获得刻度线长度 plot_theme< - 函数（p）{ plyr :: defaults（p $ theme，theme_get（））} tml< - plot_theme（m）$ axis.ticks.length＃刻度线长度 ticks $ grobs [[2]] $ y ＃交换标记标记'margin和justify ticks $ grobs [[1]]< - vinvert_title_grob（ticks $ grobs [[1]]） ＃将勾号和勾号标签放回xaxis xaxis $ children [[2]] ＃将轴添加到g1 $的顶部b $ b g1 g1< - gtable_add_grob（g1，xaxis，pp $ t + 1，pp $ l，pp $ t + 1，pp $ r，clip =off，name =axis-t） ＃删除原来的x轴和xlab g1 = g1 [-c（9,10），] ＃绘制它 grid.newpage（） grid.draw （g1） I have created a graph in ggplot2 that looks like this:depth = c(1.6,2.6,3.6, 4.6,5.6,6.6,7.6,8.6) ri <- c(0.790143779,1.485888068,2.682375391,1.728120227,0.948414515,71.43308158,4.416120653,0.125458801)df = data.frame(depth,ri) library(ggplot2)m <- qplot(ri, depth, data=df)m + scale_x_log10("Richardson Number",breaks = c(0.1,0.25,0.5,1,5,10, 50)) + scale_y_reverse("Depth (m)")This is the output:What I am trying to do is have the x-axis along the top, and also include a geom_line like the one I have added here (manually in Paint).I understand that it is difficult to get ggplot2 to move the axes around, and I have tried to reproduce this graph in ggvis but I am unable to get the log10 scale I need. Is there any way I can use R to create the graph I am aiming for? 解决方案 Updated solutionAs of ggplot 2.2.0, axes can be drawn on the top of the panel (and/or on the right of the panel)library(ggplot2)depth = c(1.6,2.6,3.6, 4.6,5.6,6.6,7.6,8.6) ri <- c(0.790143779,1.485888068,2.682375391,1.728120227,0.948414515,71.43308158,4.416120653,0.125458801)df = data.frame(depth,ri) m <- qplot(ri, depth, data=df) + scale_x_log10("Richardson Number",breaks = c(0.1,0.25,0.5,1,5,10, 50), position = "top") + scale_y_reverse("Depth (m)")+ geom_path()Original solution The original, after a little updating to ggplot version2.2.0. Axes can be moved around using gtable functions. Adapting code from @Walter's answer here, the basic idea is to: get the axis (the axis text and the tick marks); reverse the axis text and tick marks; add a new row to the gtable layout immediately above the plot panel; insert the modified axis into the new row.library(ggplot2)library(gtable)library(grid)depth = c(1.6,2.6,3.6, 4.6,5.6,6.6,7.6,8.6) ri <- c(0.790143779,1.485888068,2.682375391,1.728120227,0.948414515,71.43308158,4.416120653,0.125458801)df = data.frame(depth,ri) m <- qplot(ri, depth, data=df) + scale_x_log10("Richardson Number",breaks = c(0.1,0.25,0.5,1,5,10, 50)) + scale_y_reverse("Depth (m)")+ geom_path()# Get ggplot grobg1 <- ggplotGrob(m) ## Get the position of the plot panel in g1pp <- c(subset(g1$layout, name == "panel", se = t:r))# Title grobs have margins. # The margins need to be swapped.# Function to swap margins - # taken from the cowplot package:# https://github.com/wilkelab/cowplot/blob/master/R/switch_axis.Rvinvert_title_grob <- function(grob) { heights <- grob$heights grob$heights[1] <- heights[3] grob$heights[3] <- heights[1] grob$vp[[1]]$layout$heights[1] <- heights[3] grob$vp[[1]]$layout$heights[3] <- heights[1] grob$children[[1]]$hjust <- 1 - grob$children[[1]]$hjust grob$children[[1]]$vjust <- 1 - grob$children[[1]]$vjust grob$children[[1]]$y <- unit(1, "npc") - grob$children[[1]]$y grob}# Get xlab and swap marginsindex <- which(g1$layout$name == "xlab-b")xlab <- g1$grobs[[index]]xlab <- vinvert_title_grob(xlab)# Put xlab at the top of g1g1 <- gtable_add_rows(g1, g1$heights[g1$layout[index, ]$t], pp$t-1)g1 <- gtable_add_grob(g1, xlab, pp$t, pp$l, pp$t, pp$r, clip = "off", name="topxlab")# Get x axis (axis line, tick marks and tick mark labels)index <- which(g1$layout$name == "axis-b")xaxis <- g1$grobs[[index]]# Swap axis ticks and tick mark labelsticks <- xaxis$children[[2]]ticks$heights <- rev(ticks$heights)ticks$grobs <- rev(ticks$grobs)# Move tick marks # Get tick mark length plot_theme <- function(p) { plyr::defaults(p$theme, theme_get()) } tml <- plot_theme(m)$axis.ticks.length # Tick mark lengthticks$grobs[[2]]$y <- ticks$grobs[[2]]$y - unit(1, "npc") + tml# Swap tick mark labels' margins and justificationsticks$grobs[[1]] <- vinvert_title_grob(ticks$grobs[[1]])# Put ticks and tick mark labels back into xaxisxaxis$children[[2]] <- ticks# Add axis to top of g1g1 <- gtable_add_rows(g1, g1$heights[g1$layout[index, ]$t], pp$t)g1 <- gtable_add_grob(g1, xaxis, pp$t+1, pp$l, pp$t+1, pp$r, clip = "off", name = "axis-t")# Remove original x axis and xlabg1 = g1[-c(9,10), ]# Draw itgrid.newpage()grid.draw(g1) 这篇关于Log10顶部和Y轴geom_line上的x轴的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持！