[R] how to edit my R codes into a efficient way

[Petr Klasterecky]

Xuhong Zhu napsal(a):
> Hello, Everyone,
> 
> I am a student an a new learner of R and I am trying to do my homework
> in R. I have 10 files need to be read and process seperately. I really
> want to write the codes into something like "macro" to save the lines
> instead of repeating 10 times of similar work.
> 
> The following is part of my codes and I only extracted three lines for
> each repeating section.
> 
> 
> data.1 <- read.csv("http://pegasus.cc.ucf.edu/~xsu/CLASS/STA6704/pat1.csv",
> header = TRUE, sep = ",", quote = "",
> fill = TRUE);
> data.2 <- read.csv("http://pegasus.cc.ucf.edu/~xsu/CLASS/STA6704/pat3.csv",
> header = TRUE, sep = ",", quote = "",
> fill = TRUE);
> data.3 <- read.csv("http://pegasus.cc.ucf.edu/~xsu/CLASS/STA6704/pat4.csv",
> header = TRUE, sep = ",", quote = "",
> fill = TRUE);
> 
> baby.1 <- data.frame(cuff=data.1$avg_value,
> time=seq(1,dim(data.1)[1]), patient=rep(1, dim(data.1)[1]))
> baby.2 <- data.frame(cuff=data.2$avg_value,
> time=seq(1,dim(data.2)[1]), patient=rep(3, dim(data.2)[1]))
> baby.3 <- data.frame(cuff=data.3$avg_value,
> time=seq(1,dim(data.3)[1]), patient=rep(4, dim(data.3)[1]))
> 
> 
> I also tried the codes below but it doesn't work.
> 
> for(n in 1:10){
> mm <- data.frame(cuff=paste("data",n, sep=".")$avg_value,
> time=seq(1,dim(paste("data",n, sep="."))[1]),
> patient=rep(1,paste("data",n, sep="."))[1]))
> assign(paste("baby",n,sep="."), mm)}

This cannot work since paste() gives you quoted character output while 
functions like data.frame() etc expect a name of some R object.

You can use paste when reading individual csv files:
for(n in 1:10){
mydata <- read.csv(file=paste('...STA6704/pat',n,'.csv',sep=""), header 
= TRUE, sep = ",", quote = "", fill = TRUE)
#
... further lines to process mydata ...
}

A faster way of computing would involve reading the individual files 
into a list of dataframes and using lapply() on that list rather than 
processing the data inside the loop.

Petr

> Xuhong
> 
-- 
Petr Klasterecky
Dept. of Probability and Statistics
Charles University in Prague
Czech Republic