2013-03-16 64 views
3

功能scale_y_continuous(expand=c(0,0))去除利润率是给我的错误:试图在GGPLOT2可视化

Error: Discrete value supplied to continuous scale 

我真不明白这是什么意思。该功能被称为scale_y_continuous,而不是scale_y_discreteHere's my data

require(ggplot2) # ggplot 
gp <- read.csv("Retail_Gas_Prices.csv") 
gp$Date <- as.Date(substr(gp$Date, 1, 10), "%m/%d/%Y") 
gp_melted <- melt(gp, id = "Date") 

gas_ml_plot <- ggplot(subset(gp_melted, variable != "Weekly.US"), 
       aes(Date, value, colour = variable)) + 
       geom_line() + ggtitle("Retail Gas Prices In The US") + 
       theme(axis.title.x = element_blank()) + 
       ylab("Cost in Dollars") + 
       theme(axis.ticks = element_blank()) + 
       labs(colour = "US Region") + 
       scale_color_discrete(labels = c("East Coast", "Midwest", 
       "Gulf Coast", "Rocky Mountain", "West Coast")) + 
       theme(legend.background = element_blank()) + 
       theme(legend.position = c(0, 1)) + 
       theme(legend.justification = c(0, 1)) + 
       scale_y_continuous(expand = c(0, 0)) + 
       scale_x_discrete(expand = c(0, 0)) 

回答

4

该错误消息是错误的,所述问题是日期对象传递到一个离散的x缩放,该离散值到一个连续的尺度提供。解决方案是使用正确的x尺度;由于x变量是日期,因此使用scale_x_date(expand = c(0, 0))代替scale_x_discrete(expand = c(0, 0))

我已在https://github.com/hadley/ggplot2/issues/783

+0

提交了一个描述此问题的错误报告谢谢Ista!我们可以将你编译成ggplot2的某种人类代码扩展,以便我可以得到有用的错误吗? :P – KFunk 2013-03-17 00:27:22

+0

哈哈,@KFunk我想不出让它自动化的方法,只是继续张贴在SO上,我会尽我所能。 – Ista 2013-03-17 00:30:13