Question

library(gridExtra)
library(grid)
library(tidyquant) # theme_tq()
library(ggplot2)

我有以下数据：

head(training,20)
   gender_dom marital     race edu_level rental inc_level cluster_Kproto
1           F  single   others jrcollege rented      high              1
2           M married hispanic  postgrad leased    medium              3
3           M   other hispanic  highschl rented      high              1
6           M   other hispanic  postgrad rented    medium              3
7           M married    black doctorate leased    medium              3
8           M married hispanic jrcollege  owned      high              2
10          F  single   others  graduate rented       low              3
12          F married    asian  highschl  owned    medium              3
14          M  single hispanic  graduate leased      high              1
16          F married    white  postgrad rented    medium              3
18          M   other   others  postgrad leased    medium              3
22          F  single   others  graduate leased      high              2
23          M  single    asian doctorate leased    medium              3
25          F   other    white  highschl rented    medium              3
26          M   other    asian jrcollege leased       low              3
27          F  single    white jrcollege leased    medium              3
28          M married    asian doctorate rented       low              3
29          F   other    white  highschl rented      high              1
30          F  single hispanic jrcollege leased      high              2
31          F   other    asian jrcollege  owned       low              3

# Make variables into factors
factor_vars <- c('gender_dom','marital','race','edu_level','rental','inc_level','cluster_Kproto')
training[factor_vars] <- lapply(training[factor_vars], function(x) as.factor(x))
str(training)

library(tidyquant) # theme_tq()
par(ask=F) # to remove hit enter for each plot

(p1 <- training %>% ggplot(aes(x = race, fill = cluster_Kproto)) + geom_bar(position = position_dodge(width = 0.8), 
    width = 0.7, alpha = 0.8) + scale_fill_manual(values = palette_light()) + 
    theme_tq() + theme(legend.position = "right") + guides(fill = guide_legend("Cluster")) + 
    ggtitle("Cluster Distribution by Race") + theme(plot.title = element_text(hjust = 0.5)) + 
    theme(axis.text.x = element_text(angle = 45, vjust = 1, hjust = 1)) + 
    labs(x = "", fill = ""))

(p2 <- training %>% ggplot(aes(x = edu_level, fill = cluster_Kproto)) + geom_bar(position = position_dodge(width = 0.8), 
    width = 0.7, alpha = 0.8) + scale_fill_manual(values = palette_light()) + 
    theme_tq() + theme(legend.position = "right") + guides(fill = guide_legend("Cluster")) + 
    ggtitle("Cluster Distribution by Education level") + theme(plot.title = element_text(hjust = 0.5)) + 
    theme(axis.text.x = element_text(angle = 45, vjust = 1, hjust = 1)) + 
    labs(x = "", fill = ""))

(p3 <- training %>% ggplot(aes(x = marital, fill = cluster_Kproto)) + geom_bar(position = position_dodge(width = 0.8), 
    width = 0.7, alpha = 0.8) + scale_fill_manual(values = palette_light()) + 
    theme_tq() + theme(legend.position = "right") + guides(fill = guide_legend("Cluster")) + 
    ggtitle("Cluster Distribution by Marital") + theme(plot.title = element_text(hjust = 0.5)) + 
    theme(axis.text.x = element_text(angle = 45, vjust = 1, hjust = 1)) + 
    labs(x = "", fill = ""))

grid.arrange(p1, p2,p3,ncol = 2,nrow=2)

我想要做的是将cluster_kproto的所有分类变量按照上面的图中的几个变量进行迭代绘制，包括绘图标题也来自ggplot函数的x值，并且都是网格这样安排，如果我的分类变量没有变化，我仍然得到网格排列的情节，而不需要为每个分类变量进行手动绘图，因为我现在正在进行。

在这里需要一些帮助!!!!!

Answer 1

创建自定义函数plotMyData，然后循环遍历要绘制的变量：

plotMyData <- function(varName, inputData) {
    ggplot(inputData, aes(x = get(varName), fill = cluster_Kproto)) + 
        geom_bar(position = position_dodge(width = 0.8), 
                 width = 0.7, alpha = 0.8) + 
        scale_fill_manual(values = palette_light()) + 
        ggtitle(paste("Cluster Distribution by", varName)) + 
        labs(x = "", fill = "") +
        guides(fill = guide_legend("Cluster")) + 
        theme_tq() + 
        theme(legend.position = "right") + 
        theme(plot.title = element_text(hjust = 0.5)) + 
        theme(axis.text.x = element_text(angle = 45, vjust = 1, hjust = 1))
}

plotList <- list()
varToPlot <- c("race", "edu_level", "marital")

for(i in varToPlot) {
    plotList[[i]] <- plotMyData(i, training)
}
do.call("grid.arrange", c(plotList, ncol = 2, nrow = 2))

r ggplot如何迭代地绘制df中的所有字符变量并动态地在网格结构中绘制它们

1 个答案: