斜線上的boxplot的x標籤R

Question

我試圖尋找一種方法來傾斜我的boxplot，但從未降落到名稱的正確位置。

我有這個數據框Accuracy不同精度的每個模型，我想要顯示一個boxplot的變異性。

sample iteration poly_eps.001_C1 poly_eps.01_C1 poly_eps.1_C1 poly_eps.001_C10 poly_eps.01_C10 
(int)  (int)   (dbl)   (dbl)   (dbl)   (dbl)   (dbl) 
    1   1  0.9841270  0.9841270  0.9761905  0.9761905  0.9761905 
    1   2  0.9680000  0.9680000  0.9520000  0.9600000  0.9600000 
    1   3  0.9523810  0.9523810  0.9603175  0.9365079  0.9365079 
    1   4  0.9600000  0.9600000  0.9600000  0.9520000  0.9520000 
    1   5  0.9523810  0.9523810  0.9682540  0.9444444  0.9444444 

然後我的代碼

 # algorithms helpsvme only getting the desired features for the boxplot 
    algorithms = names(accuracies_table[-c(1:2)]) 

    boxplot(accuracies_table[algorithms], col = color , 
     main = "Polynomial KSVM Models", ylab = "Accuracy", 
     las=2,ylim = c(0.90,1)) 

可能有人幫助我傾向於x軸的標籤？或者如何在ggplot2中完成？

Answer 1

試試這個：

library(ggplot2) 
library(reshape2) 
df <- melt(accuracies_table[algorithms]) 
ggplot(df, aes(variable, value, fill=variable)) + geom_boxplot() + 
     ggtitle("Polynomial KSVM Models") + ylab("Accuracy") + 
    theme(axis.text.x = element_text(angle = 45, size=15, hjust = 1))