R help - Sep 2008 - looking for a better way to code a bar graph

I've just used r a little, so  this might be a trivial question but 

 

I have a small  data file

part,   vs/total         ,ps/total,    gpgpu,   date

...       ...               ...             ...           ..

 

that contains the percent that certain features contribute to a part and
the date the part was built 

 

 

I'd like to do stacked bar graphs one bar per part showing the percents 

and  I'd like tospace the bars based on the date released

 

I tried 

 

area <- read.csv("c:/talks/pact2008/g1.csv");

dates <-strptime(as.character(area$date), "%m/%d/%Y")

 

##  hack to figure out the amount to shift the bars 

l  <- length(as.POSIXct(dates))

diff <- as.numeric(difftime(dates[2:l], dates[1:(l-1)]))

diff <- rev(diff/max(diff))

 

colors <- gray.colors(3)

 

 

data <- c(area$ps.total[1:5], area$vs.total[1:5], area$gpgpu[1:5])

 

data <- 100*t(matrix(data, 5, 3))

 

## the .6 makes the bars look a little lighter, I did ths by eye

barplot(data,  space=diff+.6)

 

 

 

This works giving me the graph I want, but it seems so clunky, what is a
better way?

 

 

I do get a warning 

 

Warning messages:

1: In space + width :

  longer object length is not a multiple of shorter object length

2: In space + width :

  longer object length is not a multiple of shorter object length
> 
 

making me think I'm not understanding how space works.  I don't
understand how to set width, what kind of units would it be in?

 


	[[alternative HTML version deleted]]

Hi Norman,

This is pretty easy in ggplot2 (if you adjust your data a little):

install.packages("ggplot2")
library(ggplot2)

days <- seq(Sys.Date(), len=100, by="1 day")

df <- data.frame(
   date = rep(days, each = 3),
   trt = rep(c("A", "B", "C"), 50),
   price = runif(150)
)

qplot(date, price, data=df, geom="bar", stat="identity",
fill=trt)
qplot(date, price, data=df, geom="bar", stat="identity",
fill=trt,
position = "fill")
qplot(date, price, data=df, geom="bar", stat="identity",
fill=trt,
position = "fill") + scale_fill_grey()

Hadley

On Fri, Sep 26, 2008 at 8:36 AM, Rubin, Norman <Norman.Rubin at amd.com>
wrote:
> I've just used r a little, so  this might be a trivial question but
>
>
>
> I have a small  data file
>
> part,   vs/total         ,ps/total,    gpgpu,   date
>
> ...       ...               ...             ...           ..
>
>
>
> that contains the percent that certain features contribute to a part and
> the date the part was built
>
>
>
>
>
> I'd like to do stacked bar graphs one bar per part showing the percents
>
> and  I'd like tospace the bars based on the date released
>
>
>
> I tried
>
>
>
> area <- read.csv("c:/talks/pact2008/g1.csv");
>
> dates <-strptime(as.character(area$date), "%m/%d/%Y")
>
>
>
> ##  hack to figure out the amount to shift the bars
>
> l  <- length(as.POSIXct(dates))
>
> diff <- as.numeric(difftime(dates[2:l], dates[1:(l-1)]))
>
> diff <- rev(diff/max(diff))
>
>
>
> colors <- gray.colors(3)
>
>
>
>
>
> data <- c(area$ps.total[1:5], area$vs.total[1:5], area$gpgpu[1:5])
>
>
>
> data <- 100*t(matrix(data, 5, 3))
>
>
>
> ## the .6 makes the bars look a little lighter, I did ths by eye
>
> barplot(data,  space=diff+.6)
>
>
>
>
>
>
>
> This works giving me the graph I want, but it seems so clunky, what is a
> better way?
>
>
>
>
>
> I do get a warning
>
>
>
> Warning messages:
>
> 1: In space + width :
>
>  longer object length is not a multiple of shorter object length
>
> 2: In space + width :
>
>  longer object length is not a multiple of shorter object length
>
>>
>
>
>
> making me think I'm not understanding how space works.  I don't
> understand how to set width, what kind of units would it be in?
>
>
>
>
>        [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>


-- 
http://had.co.nz/