Producing multiple R graph files in a loop

[Li Xiaobo]

hello!

 

I have a question you could give me some tips:

 

Assume I have 3 categories of data: and I would like to create a loop to have a scatter plot for each and the Title of the graph should contains “Category of plain”; I would like the code to be flexible so that for each category scatter plot, the program can give the name automatically. And also for the final graph, I would like to save into 3 files which has good visual effect in PPT presentation with the name of category: plain.bmp, river.bmp, and valley.bmp.

 

 

 

Thanks a lot!

Jack

 

category	x	y
plain	0.244655	1.582322
plain	0.227968	1.853089
plain	0.544354	1.69098
plain	0.430162	1.041818
plain	0.58876	1.541161
plain	0.591491	1.190874
plain	0.326573	1.432451
plain	0.377955	1.763906
plain	0.252851	1.10793
river	2.669902	3.511024
river	2.534036	3.722374
river	2.522919	3.857419
river	2.227972	3.451399
river	2.270864	3.445889
river	2.796426	3.734166
river	2.860347	3.398596
river	2.575079	3.029064
river	2.38188	3.494636
river	2.159525	3.229826
river	2.999258	3.956068
valley	10.08767	13.88812
valley	10.97654	13.39109
valley	10.34369	13.60074
valley	10.82515	13.53325
valley	10.91398	13.67478
valley	10.84971	13.41532
valley	10.29878	13.91667
valley	10.26964	13.77659
valley	10.98827	13.46049
valley	10.49848	13.84646
valley	10.52983	13.78469
valley	10.27991	13.99628
valley	10.67307	13.07919
valley	10.9463	13.54925
valley	10.92937	13.28916
valley	10.93894	13.44915
valley	10.34638	13.92315
valley	10.10787	13.88503

  

[Raphael Saldanha]

Hi,

Assuming that you already have your data frame in R, you can try begin with something like this:

categories <- c("plain","river","valley")

for (i in categories)
{
  plot_data <- subset(data_frame, category == i)
  plot(plot_data)
}

Where data_frame represents your data frame.

--
You received this message because you are subscribed to the ggplot2 mailing list.
Please provide a reproducible example: http://gist.github.com/270442
 
To post: email ggp...@googlegroups.com
To unsubscribe: email ggplot2+u...@googlegroups.com
More options: http://groups.google.com/group/ggplot2

--
Atenciosamente,

Raphael Saldanha
saldanha...@gmail.com

[Raphael Saldanha]

Here is a more complete solution:

library(ggplot2)

info <- read.csv("info.csv",header=TRUE)

categories <- unique(info$category)

for (i in categories)
{
  plot_data <- subset(info, category == i)
  qplot(x=x,y=y,data=plot_data, main=paste("Category of ",i,sep=""))
}

Where info is your data frame, loaded from a CSV file.

[Ben Bond-Lamberty]

A final step, inside the loop, would be

>  ggsave(paste(i,"pdf",sep="."))   # or jpg or whatever

Ben

[Raphael Saldanha]

Thanks Ben, I have forgot this. So, the code could be something like this:

library(ggplot2)

info <- read.csv("info.csv",header=TRUE)

categories <- unique(info$category)

for (i in categories)
{
  plot_data <- subset(info, category == i)

  plot <- qplot(x=x,y=y,data=plot_data, main=paste("Category of ",i,sep=""))
  ggsave(filename=paste("Category of ",i,".pdf",sep=""), plot=plot)
}

Before saving the plots, remember to set your working directory. Take a look on ?setwd

--

You received this message because you are subscribed to the ggplot2 mailing list.
Please provide a reproducible example: http://gist.github.com/270442
 
To post: email ggp...@googlegroups.com
To unsubscribe: email ggplot2+u...@googlegroups.com
More options: http://groups.google.com/group/ggplot2

[Matthew Krachey]

If you have collaborators that use a lot of command line tools, such as Mac and Linux users, or if you automate importing or other operations on files (and folders) in R, you may want to save them a little bit of a headache and not put spaces in the file/folder name. It's a good habit to get into.

Matthew