添加分组变量名称以使用 group_nest 和 map 绘制答案

【问题标题】：Adding grouped variable name to plot with group_nest and map添加分组变量名称以使用 group_nest 和 map 绘制
【发布时间】：2019-10-14 20:08:42
【问题描述】：

这可能是一个语法问题。这是用一些虚拟数据重新创建的问题。

我正在处理一个更大的数据集，我想执行一个 group_nest，然后创建一些带有组标题的图。一切顺利，直到标题重复。

library(tidyverse)

N <- 30
df <- tibble(type = rep(c("small","medium","high"), each=N/3),
             dummy = rep(c(1,5,10),each=10),
             xvals = rep(1:10,3),
             A = rnorm(N)*dummy,
             B = rnorm(N)*dummy,
             C = rnorm(N)*dummy) %>%
  mutate(type = factor(type, levels=c("small","medium","high"))) %>%
  pivot_longer(cols=-c(type,xvals), names_to="metric", values_to = "value") %>%
  group_by(type) %>%
  group_nest(.key="data") %>%
  mutate(gplot = map(data, ~ggplot(data=.x, aes(x=xvals,y=value))+
                       geom_point() +
                       facet_grid(rows=vars(metric)) +
                       ggtitle(paste0(type) ## THIS IS WHERE THE PROBLEM IS ##
                      )))

#view the plots
df$gplot[[1]] #has title "small"
df$gplot[[2]] #has title "small", should be "medium"
df$gplot[[3]] #has title "small", should be "large"

这是我得到的：

这就是我想要的：

【问题讨论】：

标签： r ggplot2 dplyr purrr

【解决方案1】：

也许改用map2？

df <- tibble(type = rep(c("small","medium","high"), each=N/3),
                         dummy = rep(c(1,5,10),each=10),
                         xvals = rep(1:10,3),
                         A = rnorm(N)*dummy,
                         B = rnorm(N)*dummy,
                         C = rnorm(N)*dummy) %>%
    mutate(type = factor(type, levels=c("small","medium","high"))) %>%
    pivot_longer(cols=-c(type,xvals), names_to="metric", values_to = "value") %>%
    group_by(type) %>%
    group_nest(.key="data") %>%
    mutate(gplot = map2(data,type, ~ggplot(data=.x, aes(x=xvals,y=value))+
                                            geom_point() +
                                            facet_grid(rows=vars(metric)) +
                                            ggtitle(.y ## THIS IS WHERE THE PROBLEM IS ##
                                            )))

【讨论】：