ggplot2包|盒箱图

专题介绍：R是一种广泛用于数据分析和统计计算的强大语言，于上世纪90年代开始发展起来。得益于全世界众多爱好者的无尽努力，大家继而开发出了一种基于R但优于R基本文本编辑器的R Studio（用户的界面体验更好）。也正是由于全世界越来越多的数据科学社区和用户对R包的慷慨贡献，让R语言在全球范围内越来越流行。其中一些R包，例如MASS，SparkR， ggplot2，使数据操作,可视化和计算功能越来越强大。R是用于统计分析、绘图的语言和操作环境。R是属于GNU系统的一个自由、免费、源代码开放的软件，它是一个用于统计计算和统计制图的优秀工具。R作为一种统计分析软件，是集统计分析与图形显示于一体的。它可以运行于UNIX、Windows和Macintosh的操作系统上，而且嵌入了一个非常方便实用的帮助系统，相比于其他统计分析软件，R的学术性开发比较早，适合生物学和医学等学术学科的科研人员使用。

是新朋友吗？记得先点R语言关注我哦～

《R语言实践》专栏·第7篇

文 | RUser

3953字 |6分钟阅读

【R语言】开通了R语言群，大家相互学习和交流，请扫描下方二维码，备注：姓名-R群，我会邀请你入群，一起进步和成长。

伙伴们，我最近在重温李宏毅老师的机器学课程，并且做了一份学习笔记。你若是想学习和应用机器学习，强烈推荐你进入数据科学与人工智能公众号，查看我的学习笔记，亮点是1）点击阅读原文，就可以在手机上非常方便地观看课程视频；2）加入配套的机器学习群，参与讨论和交流，让我们一起来学习和应用机器学习。

我们先回顾一下，我已经总结和分享了ggplot2包绘制折线图、面积图、柱形图、散点图和气泡图，相关文章如下：

本文介绍ggplot2包绘制盒箱图。盒箱图可用于表示x轴是离散变量，y轴是连续变量的分布情况。我们逐步绘制盒箱图，步骤如下：

第一步：R包管理

1if (!require("pacman")) {
2  install.packages("pacman")
3  require("pacman")
4}
5p_load(datasets, ggplot2, ggthemes, grid, dplyr, RColorBrewer)

第二步：数据加载和理解

1data(airquality)
2airquality <- airquality %>%
3  mutate(Month = factor(Month,
4                        labels = c("May", "Jun", "Jul", "Aug", "Sep")))
5str(airquality)
6summary(airquality)
7head(airquality)

第三步：逐步绘制盒箱图

1）基本盒箱图

p1 <-  ggplot(airquality, aes(x = Month, y = Ozone)) +  geom_boxplot()p1

2）定义x轴标签

p2 <- p1 + scale_x_discrete(name = "Month") +  scale_y_continuous(name = "Mean ozone in parts per billion")p2

3）调整y轴刻度

p3 <- p1 + scale_x_discrete(name = "Month") +  scale_y_continuous(name = "Mean ozone innparts per billion",                   breaks = seq(0, 175, 25), limits = c(0, 175))p3

4）添加标题

p4 <- p3 +  labs(title = "Frequency histogram of mean ozone",       subtitle = "Source: New York State Department of Conservation")p4

5）修改盒箱的颜色

fill <- "#4271AE"line <- "#1F3552"p5 <-  ggplot(airquality, aes(x = Month, y = Ozone)) +  geom_boxplot(fill = fill, colour = line, alpha = 0.7,               outlier.colour = "#1F3552",                outlier.shape = 20,               outlier.size = 2) +  scale_y_continuous(name = "Mean ozone innparts per billion",                     breaks = seq(0, 175, 25), limits = c(0, 175)) +  scale_x_discrete(name = "Month") +  labs(title = "Frequency histogram of mean ozone",       subtitle = "Source: New York State Department of Conservation")p5

6）字体设置

windowsFonts(  # 中文字体  lishu = windowsFont(family = "LiSu"),            # 隶书  yahei = windowsFont(family = "Microsoft YaHei"), # 微软雅黑  xinwei = windowsFont(family = "STXingwei"),      # 华文新魏  kaiti = windowsFont(family = "KaiTi"),           # 楷体  heiti = windowsFont(family = "SimHei"),          # 黑体  # 英文字体  arial = windowsFont(family = "Arial"),             # Arial字体  newman = windowsFont(family = "Times New Roman"),  #Times New Roman字体     hand = windowsFont(family = "Lucida Calligraphy"), # Lucida手写体  Helvetica = windowsFont(family = "Helvetica")      # 印刷体)

7）主题装饰

主题一：经济学杂志

p6_1 <- ggplot(airquality, aes(x = Month, y = Ozone)) +  geom_boxplot(fill = fill, colour = line) +  scale_y_continuous(name = "Mean ozone innparts per billion",                     breaks = seq(0, 175, 25), limits = c(0, 175)) +  scale_x_discrete(name = "Month") +  labs(title = "Frequency histogram of mean ozone",       subtitle = "Source: New York State Department of Conservation") +  theme_economist() + scale_fill_economist() +  theme(axis.line.x = element_line(size = .5, colour = "black"),        axis.title = element_text(size = 12),        legend.position = "bottom",        legend.direction = "horizontal",        legend.box = "horizontal",        legend.text = element_text(size = 10),        text = element_text(family = "yahei"),        plot.title = element_text(family = "yahei"))p6_1

主题二：简洁主题

p6_2 <- ggplot(airquality, aes(x = Month, y = Ozone)) +  geom_boxplot(colour = "black", fill = "#56B4E9") +  scale_y_continuous(name = "Mean ozone innparts per billion",                     breaks = seq(0, 175, 25), limits = c(0, 175)) +  scale_x_discrete(name = "Month") +  labs(title = "Frequency histogram of mean ozone",       subtitle = "Source: New York State Department of Conservation") +  theme(axis.line.x = element_line(size = 1, colour = "black"),        axis.line.y = element_line(size = 1, colour = "black"),        axis.text.x = element_text(colour = "black", size = 10),        axis.text.y = element_text(colour = "black", size = 10),        legend.position = "bottom",        legend.direction = "horizontal",        legend.box = "horizontal",        legend.key = element_blank(),        panel.grid.major = element_blank(),        panel.grid.minor = element_blank(),        panel.background = element_blank(),        plot.title = element_text(family = "Helvetica"),        text = element_text(family = "Helvetica"))p6_2

8）分组观察

# 数据处理airquality_trimmed <- airquality %>%  filter(Month %in% c("Jul", "Aug", "Sep")) %>%  mutate(Temp.f = factor(ifelse(Temp > mean(Temp), 1, 0),                         labels = c("Low temp ", "High temp ")))

8.1）分面板

p8_1 <- ggplot(airquality_trimmed, aes(x = Month, y = Ozone)) +  geom_boxplot(fill = fill, colour = line,               alpha = 0.7) +  scale_y_continuous(name = "Mean ozone innparts per billion",                     breaks = seq(0, 175, 50), limits = c(0, 175)) +  scale_x_discrete(name = "Month") +  labs(title = "Frequency histogram of mean ozone",       subtitle = "Source: New York State Department of Conservation") +  theme_bw() +  theme(plot.title = element_text(size = 14, family = "Tahoma", face = "bold"),        panel.border = element_rect(colour = "black", fill = NA, size = .5),        text = element_text(size = 12, family = "Tahoma"),        axis.title = element_text(face = "bold"),        axis.text.x = element_text(size = 11)) +  facet_grid(. ~ Temp.f)p8_1

8.2）同一个图形表示

p8_2 <- ggplot(airquality_trimmed, aes(x = Month, y = Ozone, fill = Temp.f)) +  geom_boxplot(alpha = 0.7) +  scale_y_continuous(name = "Mean ozone innparts per billion",                     breaks = seq(0, 175, 25), limits = c(0, 175)) +  scale_x_discrete(name = "Month") +  labs(title = "Frequency histogram of mean ozone",       subtitle = "Source: New York State Department of Conservation") +  theme_bw() +  theme(plot.title = element_text(size = 14, family = "Tahoma", face = "bold"),        panel.border = element_rect(colour = "black", fill = NA, size = .5),        text = element_text(size = 12, family = "Tahoma"),        axis.title = element_text(face = "bold"),        axis.text.x = element_text(size = 11),        legend.position = "bottom") +  scale_fill_brewer(palette = "Accent") +  labs(fill = "Temperature ")p8_2

关于盒箱图，有什么问题，请留言。

附录：完整代码

  1##################
  2#盒箱图
  3##################
  4
  5# 1 R包管理
  6if (!require("pacman")) {
  7  install.packages("pacman")
  8  require("pacman")
  9}
 10p_load(datasets, ggplot2, ggthemes, grid, dplyr, RColorBrewer)
 11
 12# 2 数据加载和理解
 13data(airquality)
 14airquality <- airquality %>%
 15  mutate(Month = factor(Month,
 16                        labels = c("May", "Jun", "Jul", "Aug", "Sep")))
 17str(airquality)
 18summary(airquality)
 19head(airquality)
 20
 21# 3 逐步绘制盒箱图
 22# 1）基本盒箱图
 23p1 <-  ggplot(airquality, aes(x = Month, y = Ozone)) +
 24  geom_boxplot()
 25p1
 26
 27# 2) 定义x轴标签
 28p2 <- p1 + scale_x_discrete(name = "Month") +
 29  scale_y_continuous(name = "Mean ozone in parts per billion")
 30p2
 31
 32# 3) 调整y轴的刻度
 33p3 <- p1 + scale_x_discrete(name = "Month") +
 34  scale_y_continuous(name = "Mean ozone innparts per billion",
 35                   breaks = seq(0, 175, 25), limits = c(0, 175))
 36p3
 37
 38# 4) 添加标题
 39p4 <- p3 +
 40  labs(title = "Frequency histogram of mean ozone",
 41       subtitle = "Source: New York State Department of Conservation")
 42p4
 43
 44# 5) 修改盒箱的颜色
 45fill <- "#4271AE"
 46line <- "#1F3552"
 47p5 <-  ggplot(airquality, aes(x = Month, y = Ozone)) +
 48  geom_boxplot(fill = fill, colour = line, alpha = 0.7,
 49               outlier.colour = "#1F3552", 
 50               outlier.shape = 20,
 51               outlier.size = 2) +
 52  scale_y_continuous(name = "Mean ozone innparts per billion",
 53                     breaks = seq(0, 175, 25), limits = c(0, 175)) +
 54  scale_x_discrete(name = "Month") +
 55  labs(title = "Frequency histogram of mean ozone",
 56       subtitle = "Source: New York State Department of Conservation")
 57p5
 58# 6) 字体设置
 59windowsFonts(
 60  # 中文字体
 61  lishu = windowsFont(family = "LiSu"),            # 隶书
 62  yahei = windowsFont(family = "Microsoft YaHei"), # 微软雅黑
 63  xinwei = windowsFont(family = "STXingwei"),      # 华文新魏
 64  kaiti = windowsFont(family = "KaiTi"),           # 楷体
 65  heiti = windowsFont(family = "SimHei"),          # 黑体
 66  # 英文字体
 67  arial = windowsFont(family = "Arial"),             # Arial字体
 68  newman = windowsFont(family = "Times New Roman"),  #Times New Roman字体   
 69  hand = windowsFont(family = "Lucida Calligraphy"), # Lucida手写体
 70  Helvetica = windowsFont(family = "Helvetica")      # 印刷体
 71)
 72
 73# 7） 主题装饰
 74# 7.1）经济学杂志主题
 75p6_1 <- ggplot(airquality, aes(x = Month, y = Ozone)) +
 76  geom_boxplot(fill = fill, colour = line) +
 77  scale_y_continuous(name = "Mean ozone innparts per billion",
 78                     breaks = seq(0, 175, 25), limits = c(0, 175)) +
 79  scale_x_discrete(name = "Month") +
 80  labs(title = "Frequency histogram of mean ozone",
 81       subtitle = "Source: New York State Department of Conservation") +
 82  theme_economist() + scale_fill_economist() +
 83  theme(axis.line.x = element_line(size = .5, colour = "black"),
 84        axis.title = element_text(size = 12),
 85        legend.position = "bottom",
 86        legend.direction = "horizontal",
 87        legend.box = "horizontal",
 88        legend.text = element_text(size = 10),
 89        text = element_text(family = "yahei"),
 90        plot.title = element_text(family = "yahei"))
 91p6_1
 92# 7.2) 简洁主题
 93p6_2 <- ggplot(airquality, aes(x = Month, y = Ozone)) +
 94  geom_boxplot(colour = "black", fill = "#56B4E9") +
 95  scale_y_continuous(name = "Mean ozone innparts per billion",
 96                     breaks = seq(0, 175, 25), limits = c(0, 175)) +
 97  scale_x_discrete(name = "Month") +
 98  labs(title = "Frequency histogram of mean ozone",
 99       subtitle = "Source: New York State Department of Conservation") +
100  theme(axis.line.x = element_line(size = 1, colour = "black"),
101        axis.line.y = element_line(size = 1, colour = "black"),
102        axis.text.x = element_text(colour = "black", size = 10),
103        axis.text.y = element_text(colour = "black", size = 10),
104        legend.position = "bottom",
105        legend.direction = "horizontal",
106        legend.box = "horizontal",
107        legend.key = element_blank(),
108        panel.grid.major = element_blank(),
109        panel.grid.minor = element_blank(),
110        panel.background = element_blank(),
111        plot.title = element_text(family = "Helvetica"),
112        text = element_text(family = "Helvetica"))
113p6_2
114
115# 8) 分组观察
116# 数据处理
117airquality_trimmed <- airquality %>%
118  filter(Month %in% c("Jul", "Aug", "Sep")) %>%
119  mutate(Temp.f = factor(ifelse(Temp > mean(Temp), 1, 0),
120                         labels = c("Low temp ", "High temp ")))
121# 8.1）分面板
122p8_1 <- ggplot(airquality_trimmed, aes(x = Month, y = Ozone)) +
123  geom_boxplot(fill = fill, colour = line,
124               alpha = 0.7) +
125  scale_y_continuous(name = "Mean ozone innparts per billion",
126                     breaks = seq(0, 175, 50), limits = c(0, 175)) +
127  scale_x_discrete(name = "Month") +
128  labs(title = "Frequency histogram of mean ozone",
129       subtitle = "Source: New York State Department of Conservation") +
130  theme_bw() +
131  theme(plot.title = element_text(size = 14, family = "Tahoma", face = "bold"),
132        panel.border = element_rect(colour = "black", fill = NA, size = .5),
133        text = element_text(size = 12, family = "Tahoma"),
134        axis.title = element_text(face = "bold"),
135        axis.text.x = element_text(size = 11)) +
136  facet_grid(. ~ Temp.f)
137p8_1
138
139# 8.2) 同一个图形表示
140p8_2 <- ggplot(airquality_trimmed, aes(x = Month, y = Ozone, fill = Temp.f)) +
141  geom_boxplot(alpha = 0.7) +
142  scale_y_continuous(name = "Mean ozone innparts per billion",
143                     breaks = seq(0, 175, 25), limits = c(0, 175)) +
144  scale_x_discrete(name = "Month") +
145  labs(title = "Frequency histogram of mean ozone",
146       subtitle = "Source: New York State Department of Conservation") +
147  theme_bw() +
148  theme(plot.title = element_text(size = 14, family = "Tahoma", face = "bold"),
149        panel.border = element_rect(colour = "black", fill = NA, size = .5),
150        text = element_text(size = 12, family = "Tahoma"),
151        axis.title = element_text(face = "bold"),
152        axis.text.x = element_text(size = 11),
153        legend.position = "bottom") +
154  scale_fill_brewer(palette = "Accent") +
155  labs(fill = "Temperature ")
156p8_2
157