3变量堆叠barplot

Question

基于R目前，我有这样的代码：

### Uses the library ggplot2### 
library("ggplot2") 
library("reshape2") 

### Reads in the CSV file to be plotted ### 
plot <- read.csv("C:/Users/dam203/Desktop/Ongoing_Projects/output-1.csv") 

### Makes R recognize that the X-axis has been pre-sorted so that GGPLOT2 does not sort alphabetically. ### 
plot$Date <- factor(plot$Date, levels = plot$Date) 

### Plots the Graph ### 
ggplot(plot[which(plot$F.Sym.Onset>0),], aes(x=Date, y=F.Sym.Onset)) + geom_bar(stat="identity") + theme(axis.text.x=element_text(angle=90, hjust=1)) + ggtitle("Epidemic Pertussis Case Curve") 

这是什么CSV文件中的数据看起来像一个小样本。但是，实际CSV文件中有更多的列和行。日期，C.Sym.Onset，F.Sym.Onset，D.Sym.Onset是我现在很感兴趣：

Date C.Sym.Onset F.Sym.Onset D.Sym.Onset  Temp 
6-Jan  2    1      47 
7-Jan  1    3   2   57 
8-Jan         1   54 
9-Jan            58 
10-Jan  1          59 

由于代码的正上方，它会忽略日期没有F唯一列.Sym.Onset并绘制根据图表组织的给定日期的案例数量，而不是ggplots默认的字母顺序。

我的问题是，我该怎么做才能在垂直堆积的条形图上绘制C.Sym.onset，F.Sym.Onset和D.Sym.Onset？

这是一个图表的副本我的剧本目前从产生CSV具有完整的数据文件：

RPlot

感谢您的帮助！

Answer 1

library(ggplot2) 
library(tidyr) 
library(dplyr) 

df <- data.frame(Date = seq.Date(Sys.Date()-5, Sys.Date(), by = 'days'), 
       x = sample(c(1:3, NA), 6, replace = TRUE), 
       y = sample(c(1:3, NA), 6, replace = TRUE), 
       z = sample(c(1:3, NA), 6, replace = TRUE)) 
longdf <- gather(df, var, value, -Date) 

ggplot(longdf, aes(value, fill = var)) + 
    geom_bar() 

您需要将您的列收集到一个变量中。然后，您可以通过指定填充var来将它们绘制为堆积条形图。

Answer 2

我改写的代码，我能够创建使用用下面的代码熔化堆积条形图积：

### Uses the library ggplot2### 
library("ggplot2") 
library("reshape2") 

### Reads in the CSV file to be plotted ### 
ModelOutput <- read.csv("C:/Users/dam203/Desktop/Ongoing_Projects/output-3.csv") 

### Creates a new dataframe using only variables needed to plot the PECurve. ### 
PECurve <- data.frame("Date" = ModelOutput$Date, 
       "C.Sym.Onset" = ModelOutput$C.Sym.Onset, 
       "F.Sym.Onset" = ModelOutput$F.Sym.Onset, 
       "D.Sym.Onset" = ModelOutput$D.Sym.Onset) 

### Removes rows where value is 'NA' in C.Sym.Onset, F.Sym.Onset, and D.Sym.Onset. ### 
PECurve <- PECurve[!with(PECurve,is.na(C.Sym.Onset)& is.na(F.Sym.Onset)& is.na(D.Sym.Onset)),] 

### Makes R recognize that the X-axis has been pre-sorted so that GGPLOT2 does not sort alphabetically. ### 
PECurve$Date <- factor(PECurve$Date, levels = PECurve$Date) 

### Plots according to Day of Symptom Onset ### 
df <- melt(PECurve, .measure.vars=.(C.Sym.Onset, F.Sym.Onset, D.Sym.Onset)) 
ggplot(df, aes(x=Date, y=value, fill=variable)) + 
    geom_bar(stat="identity") + theme(axis.text.x=element_text(angle=90, hjust=1)) 

从CSV最终情节的例子是这样的：

GGPlot2