CODEWITHSUNDEEP
rbar-chart
Ruser
Ruser

Reputation: 85

R - barplot grouping columns

I have the following data which contains data from 7 combinations (rows) and 12 methods (columns).

  structure(list(Beams = structure(c(1L, 3L, 4L, 5L, 6L, 7L, 2L
), .Label = c("1 – 2", "1 – 2 – 3 – 4", "1 – 3", "1 – 4", "2 – 3", 
"2 – 4", "3 – 4"), class = "factor"), Slope...No.weight = c(75L, 
65L, 45L, 30L, 95L, 70L, 75L), Slope...W1 = c(85L, 70L, 65L, 
55L, 90L, 85L, 75L), Slope...W2 = c(80L, 65L, 65L, 50L, 90L, 
90L, 75L), Slope...W3 = c(80L, 75L, 75L, 65L, 90L, 95L, 80L), 
Average.Time...No.Weight = c(75L, 65L, 45L, 30L, 95L, 70L, 
70L), Average.Time...W1 = c(70L, 60L, 75L, 60L, 75L, 75L, 
80L), Average.Time...W2 = c(65L, 40L, 65L, 50L, 75L, 85L, 
70L), Average.Time...W3 = c(65L, 40L, 80L, 75L, 65L, 85L, 
80L), Momentum...No.weight = c(80L, 60L, 45L, 30L, 95L, 70L, 
75L), Momentum...W1 = c(85L, 75L, 60L, 55L, 95L, 90L, 80L
), Momentum...W2 = c(80L, 65L, 70L, 50L, 90L, 90L, 85L), 
Momentum...W3 = c(85L, 75L, 75L, 55L, 90L, 95L, 80L)), .Names =   c("Beams", 
"Slope...No.weight", "Slope...W1", "Slope...W2", "Slope...W3", 
"Average.Time...No.Weight", "Average.Time...W1", "Average.Time...W2", 
"Average.Time...W3", "Momentum...No.weight", "Momentum...W1", 
"Momentum...W2", "Momentum...W3"), class = "data.frame", row.names = c(NA, 
-7L))

I would like to get a barplot like the one below:

enter image description here

I've tried with 

library(RColorBrewer)
dat<-read.csv("phaser-p13-30dBm-100ms.csv")

names <- c("1-2","1-3","1-4","2-3","2-4","3-4","1-2-3-4")

barx <- 

 barplot(as.integer(dat2[,2:13]),
      beside=TRUE,
      col=brewer.pal(12,"Set3"),
      names.arg=names,          
      ylim=c(0,100),
      xlab='Combination of beams',
      ylab='Correct detection [%]')



box()
par(xpd=TRUE)
legend("top", c("Slope - No weight","Slope - W1","Slope - W2","Slope - W3","Average Time - No weight","Average Time - W1","Average Time - W2","Average Time - W3","Momentum - No weight","Momentum - W1","Momentum - W2","Momentum - W3"), fill = brewer.pal(12,"Set3"),horiz = T)

but I got this error:

Error in barplot.default(as.integer(dat2[, 2:13]), beside = TRUE, col = brewer.pal(12,  : 
  incorrect number of names

Could you find the error?

Upvotes: 2

Views: 405

Answers (3)

Anders Ellern Bilgrau
Anders Ellern Bilgrau

Reputation: 10223

This should get you started, if you wish to use base graphics and not ggplot2:

df <- as.matrix(dat[,-1])
rownames(df) <- dat[, 1]

barplot(df, beside = TRUE, las = 2)

Upvotes: 1

boshek
boshek

Reputation: 4406

I've named you dataframe df here and made use of three packages. This is not a base R solution. Given your dataset format, this is the easiest way (IMO) to do this:

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

df %>% # dataframe
  gather(variable, value, -Beams) %>% # convert to long format excluding beams column
  ggplot(aes(x=Beams, y=value, fill=variable)) + # plot the bar plot
  geom_bar(stat='identity', position='dodge')

enter image description here

Upvotes: 1

DaveRGP
DaveRGP

Reputation: 1480

Use ggplot2 package and make sure that your data is neat and ordered?

something like ggplot(dataframe, aes(colour = some_factor))) + geom_bar(aes(x=Some_variable, y=Some_other_variable))

More explict statement as to how your data matches the image would be useful.

Upvotes: 0

Related Questions