Question

我正在使用R中的barplot（）函数在堆叠条形表示下绘制来自同一绘图中的多个数据集的值，并且我注意到如果对于某个绘图我只有一个数据的数据，则不显示图例组。具有两个或更多类别（即数据集）不会引起任何问题并且图例正确显示。有没有想过是否可以强制它显示，即使只有一个类别？或者我必须添加一个虚拟类别，如果该绘图我只有一个数据集可用的数据。谢谢。

编辑：以下是我对条形图的调用方式：

barplot(bars, col = color_map[available_data], legend.text = T, 
        args.legend(bty = 'n'), ylim = my_computed_ylim, 
        xlim = my_computed_xlim, xlab = "X label", ylab = "Y label") 

a = rep(5,25) 
b = rep(10,25) 
bars = rbind(a,b) 
barplot(bars, col = seq(1,nrow(bars), by = 1), legend.text = T, 
        args.legend = c(bty = 'n')) bars = bars[-1,] barplot(bars, 
        col = 2, legend.text = T, args.legend = c(bty = 'n'))

Answer 1

当您键入bars = bars[-1,]时，会自动强制转换为矢量。为此，您应该转换回具有命名行的矩阵。

示例：

a = rep(5,25); 
b = rep(10,25); 
bars = rbind(a,b); 
barplot(bars, col = seq(1,nrow(bars), by = 1), legend.text = T, args.legend = c(bty = 'n')); 
bars = matrix(bars[-1,],nrow=1); rownames(bars)=c('b');  ### THIS IS DIFFERENT
barplot(bars, col = 2, legend.text = T, args.legend = c(bty = 'n'))

这有帮助吗？

修改

要真正看到两只野兽之间的区别，请看这个例子：

> a = rep(5,25); b = rep(10,25); bars = rbind(a,b); > bars [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9] [,10] [,11] [,12] [,13] [,14] [,15] [,16] [,17] a 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 b 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 [,18] [,19] [,20] [,21] [,22] [,23] [,24] [,25] a 5 5 5 5 5 5 5 5 b 10 10 10 10 10 10 10 10 > bars.old = bars[-1,] > bars.old [1] 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 length of 'dimnames' [1] not equal to array extent > bars.new = matrix(bars[-1,],nrow=1); rownames(bars.new)=c('b'); > bars.new [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9] [,10] [,11] [,12] [,13] [,14] [,15] [,16] [,17] b 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 [,18] [,19] [,20] [,21] [,22] [,23] [,24] [,25] b 10 10 10 10 10 10 10 10

R barplot - 即使只有一个类别，也会显示图例

1 个答案: