[R] A. Mani : Tapply

[Petr Pikal]

Hi 


> PLEASE do read the posting guide! 
****************************** 

On 25 Aug 2005 at 20:28, A Mani wrote: 

> Hello, 
>         Is it safe to use tapply when the result will be of dim 20000 
> x 10000 or more ? In my PC R crashes.  

or gave you an error message? 

I tried this 

> df<-data.frame(A=as.factor(sample(1:10000,100000,  
replace=T)),B=as.factor(sample(100000:110000,100000,  
replace=T)), num=rnorm(100000)) 
> ttt<-tapply(df$num, list(df$A,df$B), diff) 

and got 

Error: cannot allocate vector of size 390664 Kb 
In addition: Warning messages: 
1: Reached total allocation of 1000Mb: see help(memory.size)  
2: Reached total allocation of 1000Mb: see help(memory.size)  
> 

and the result with smaller data sets are 

df1<-data.frame(A=as.factor(sample(1:2,100000, 
replace=T)),B=as.factor(sample(10:11,100000, replace=T)), 
num=rnorm(100000)) 
ttt1<-tapply(df1$num, list(df1$A,df1$B), diff) 

> ttt1 
  10            11            
1 Numeric,24933 Numeric,25141 
2 Numeric,24992 Numeric,24930 


df<-data.frame(A=as.factor(sample(1:1000,100000, 
replace=T)),B=as.factor(sample(10000:11000,100000, 
replace=T)), num=rnorm(100000)) 
 ttt<-tapply(df$num, list(df$A,df$B), diff) 


> 
> ttt[1:2,1:5] 
  10000 10001 10002 10003     10004 
1 NULL  NULL  NULL  Numeric,0 NULL  
2 NULL  NULL  NULL  Numeric,0 NULL  
> 

so you are probably receiving humonguous table of NULLs, zeros  
and few nonzero entries. 

You probably need to use different approach 

Cheers 
Petr 


The code used was on a 3-col 
> data frame with two factor cols and a numeric column. The fn 
was diff 
> . data form being <A, B, Num> tapply(data$A, list(data$A, 
data$B), 
> diff) 
>  
> --  
> A. Mani 
> Member, Cal. Math. Soc 
>  
> ______________________________________________ 
> R-help at stat.math.ethz.ch mailing list 
> https://stat.ethz.ch/mailman/listinfo/r-help 
> PLEASE do read the posting guide! 
> http://www.R-project.org/posting-guide.html 

Petr Pikal
petr.pikal at precheza.cz