R,用scale_linetype_manual更改ggplot图例名称

时间:2018-05-11 13:12:29

标签: r plot ggplot2

我有一个如下所示的数据框:

> df
    Year mpft       value   type index
1   1996    2 0.033827219  solid   2.1
2   1997    2 0.133278701  solid   2.1
3   1998    2 0.261428650  solid   2.1
4   1999    2 0.394702438  solid   2.1
5   1996    3 0.019079686  solid   3.1
6   1997    3 0.074332942  solid   3.1
7   1998    3 0.149042964  solid   3.1
8   1999    3 0.227812452  solid   3.1
9   1996    4 0.009909126  solid   4.1
10  1997    4 0.026231721  solid   4.1
11  1998    4 0.052912805  solid   4.1
12  1999    4 0.086256016  solid   4.1
13  1996   17 0.017256492  solid  17.1
14  1997   17 0.079446280  solid  17.1
15  1998   17 0.166014538  solid  17.1
16  1999   17 0.316175339  solid  17.1
17  1996   18 0.080072523  solid  18.1
18  1997   18 0.313289644  solid  18.1
19  1998   18 0.629398957  solid  18.1
20  1999   18 1.024946245  solid  18.1
110 1996    2 0.031634282 dashed   2.2
21  1997    2 0.139244701 dashed   2.2
31  1998    2 0.273270126 dashed   2.2
41  1999    2 0.412409808 dashed   2.2
51  1996    3 0.019430502 dashed   3.2
61  1997    3 0.079252516 dashed   3.2
71  1998    3 0.161607337 dashed   3.2
81  1999    3 0.252595611 dashed   3.2
91  1996    4 0.009976637 dashed   4.2
101 1997    4 0.027057403 dashed   4.2
111 1998    4 0.055755671 dashed   4.2
121 1999    4 0.093064641 dashed   4.2
171 1996   18 0.061041422 dashed  18.2
181 1997   18 0.245554619 dashed  18.2
191 1998   18 0.490633135 dashed  18.2
201 1999   18 0.758070060 dashed  18.2

我正在尝试绘制数据并拥有正确的图例,到目前为止我最初尝试使用

ggplot(df,aes(x=Year,y=value, colour = factor(mpft),linetype=type)) +
      geom_line(aes(group = index), size = 1.4) +
      #scale_linetype_manual(name= "Run Type", values = unique(df$type), labels = run.type) +
      scale_color_manual(name = "PFT",
                         values = setNames(mycol[unique(df$mpft)], unique(df$mpft)),
                         labels = setNames(mynam[unique(df$mpft)], unique(df$mpft)))

哪个给了我

enter image description here

我尝试使用

添加scale_linetype_manual
ggplot(df,aes(x=Year,y=value, colour = factor(mpft),linetype=type)) +
      geom_line(aes(group = index), size = 1.4) +
      scale_linetype_manual(name= "Run Type", values = unique(df$type), labels = run.type) +
      scale_color_manual(name = "PFT",
                         values = setNames(mycol[unique(df$mpft)], unique(df$mpft)),
                         labels = setNames(mynam[unique(df$mpft)], unique(df$mpft)))

> run.type
[1] "current" "origED3"

但是我得到了

enter image description here

具有图例的正确名称,但具有不同的线型。 我错过了什么?

修改

我的数据框的dput

> dput(df)
structure(list(Year = c(1996, 1997, 1998, 1999, 1996, 1997, 1998, 
1999, 1996, 1997, 1998, 1999, 1996, 1997, 1998, 1999, 1996, 1997, 
1998, 1999, 1996, 1997, 1998, 1999, 1996, 1997, 1998, 1999, 1996, 
1997, 1998, 1999, 1996, 1997, 1998, 1999), mpft = c(2L, 2L, 2L, 
2L, 3L, 3L, 3L, 3L, 4L, 4L, 4L, 4L, 17L, 17L, 17L, 17L, 18L, 
18L, 18L, 18L, 2L, 2L, 2L, 2L, 3L, 3L, 3L, 3L, 4L, 4L, 4L, 4L, 
18L, 18L, 18L, 18L), value = c(0.0338272191848643, 0.133278701149992, 
0.261428650232716, 0.394702437670559, 0.0190796862689925, 0.0743329421068756, 
0.149042964352043, 0.227812451937011, 0.00990912614900737, 0.0262317206863519, 
0.0529128049802722, 0.0862560162908444, 0.017256491619149, 0.0794462797803606, 
0.166014537897384, 0.31617533869767, 0.0800725232220131, 0.31328964372358, 
0.629398957462415, 1.02494624459608, 0.0316342818911836, 0.139244700529005, 
0.273270126484303, 0.412409807917143, 0.0194305022713642, 0.0792525159706922, 
0.161607337403947, 0.252595610607411, 0.00997663742883768, 0.0270574028188436, 
0.0557556714277292, 0.0930646413413941, 0.0610414215913856, 0.245554619318541, 
0.490633135315979, 0.758070059865948), type = structure(c(1L, 
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 
2L, 2L, 2L), .Label = c("solid", "dashed"), class = "factor"), 
    index = structure(c(1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 3L, 3L, 
    3L, 3L, 4L, 4L, 4L, 4L, 5L, 5L, 5L, 5L, 6L, 6L, 6L, 6L, 7L, 
    7L, 7L, 7L, 8L, 8L, 8L, 8L, 9L, 9L, 9L, 9L), .Label = c("2.1", 
    "3.1", "4.1", "17.1", "18.1", "2.2", "3.2", "4.2", "18.2"
    ), class = "factor")), .Names = c("Year", "mpft", "value", 
"type", "index"), row.names = c("1", "2", "3", "4", "5", "6", 
"7", "8", "9", "10", "11", "12", "13", "14", "15", "16", "17", 
"18", "19", "20", "110", "21", "31", "41", "51", "61", "71", 
"81", "91", "101", "111", "121", "171", "181", "191", "201"), class = "data.frame")

修改

不是一个非常优雅的解决方案,而是将"dashed"替换为"22",然后使用:

ggplot(df,aes(x=Year,y=value, colour = factor(mpft),linetype=type)) +
      geom_line(aes(group = index), size = 1.4) +
      scale_linetype_manual(name= "Run Type", values = unique(as.character(df$type)), labels = run.type) +
      scale_color_manual(name = "PFT",
                         values = setNames(mycol[unique(df$mpft)], unique(df$mpft)),
                         labels = setNames(mynam[unique(df$mpft)], unique(df$mpft)))

我能够做对

enter image description here

2 个答案:

答案 0 :(得分:1)

你只是无法看到线型差异 - 如果你使图例更宽,它是可见的:

ggplot(df,aes(x=Year,y=value, colour = factor(mpft),linetype=type)) +
  geom_line(aes(group = index), size = 1.4) +
  scale_linetype_manual(name= "Run Type", values = unique(df$type), labels = run.type) +
  guides(linetype = guide_legend(keywidth = 3, keyheight = 1)) 

答案 1 :(得分:1)

有趣的问题,以及我之前没有遇到的问题。

简短回答

不同之处在于,您的第一个版本中的线型根本不是"dashed"

答案很长

在您的第一个版本中,没有为linetype美学指定任何内容,ggplot默认为scale_linetype_discrete(),而current code则为:

scale_linetype <- function(..., na.value = "blank") {
  discrete_scale("linetype", "linetype_d", linetype_pal(),
    na.value = na.value, ...)
}

因此,scale_linetype_discrete()linetype_pal函数中获取其线型值,由scales包提供(至少,这是我找到它的唯一地方):

> scales::linetype_pal()(2)
[1] "solid" "22"  

当您使用linetype在第二个版本中指定scale_linetype_manual()美学地图时,相应的current code为:

scale_linetype_manual <- function(..., values) {
  manual_scale("linetype", values, ...)
}

因此,当您明确要求c("solid", "dashed")作为绘图中的两个线型值时,ggplot会使用它们。如果不这样做,则默认值为c("solid", "22")"22"对应于与"dashed"模式不同,间隔更紧密的模式。

下面的演示,使用内置数据:

df.sample <- diamonds %>% 
  filter(cut %in% c("Fair", "Good")) %>%
  group_by(cut, clarity) %>% 
  summarise(price = mean(price / carat)) %>%
  ungroup() 

p <- ggplot(df.sample,
            aes(x = clarity, y = price, group = cut,
                linetype = cut)) +
  geom_line(size = 1) +
  guides(linetype = guide_legend(keywidth = 3, keyheight = 1)) +
  theme(legend.position = c(1, 0), legend.justification = c(1, 0))

library(gridExtra)
grid.arrange(p + 
               labs(title = "Default scale",
                    subtitle = c("values = linetype_pal()(2)")),
             p + scale_linetype_manual(values = c("solid", "dashed")) +
               labs(title = "Manual scale",
                    subtitle = "values = c('solid', 'dashed')"),
             p + scale_linetype_manual(values = c("solid", "22")) +
               labs(title = "Manual scale",
                    subtitle = "values = c('solid', '22')"),
             nrow = 1)

第三个图模仿了默认比例的行为。

plot