Skip to content

dynamic lists (data frames?)

5 messages · Tom, Vincent Goulet, Gabor Grothendieck

Original
Tom
# 
Can someone please show me what I need to get something like this to
work

for(a in c(1:5)){
	data$a<-c(a:10)
}

so that i end up with a structure
data$1<-[1,2,3,4,5,6,7,8,9,10]
data$2<-[2,3,4,5,67,8,9,10]
data$3<-[3,4,5,67,8,9,10]
data$4<-[4,5,67,8,9,10]
data$5<-[5,67,8,9,10]

thanks loads
Tom
Tom
# 
Hmm my bad,
Thanks for your replies but I think my example was a little to simple
the actual code I'm using is:


f_haardisolve<-function(v_dataset){
    #pad data to make length a power of 2
    v_dataset<-f_paddata(v_dataset)

    l_cooef<-list() #holder for cooefficents
    i_count<-1      #identity counter

    while(length(v_dataset)>0){
        #seperate odd and even points
        v_dataset<-f_splitpoints(v_dataset)

        v_dataset<-f_haarpredict(v_dataset)
        v_dataset<-f_haaruplift(v_dataset)
        
        str_idx<-paste('c',i_count,sep='')
	l_cooef
$str_idx<-v_dataset[c((length(v_dataset)/2)+1:(length(v_dataset)/2))]
#should be no wrap on the previous line 

       v_dataset<-v_dataset[c(1:length(v_dataset)/2)]

        i_count<-i_count+1
    }
    l_cooef
}

In this particular case I could probably calculate the number of
iterations and use l_cooef<-names() but more is there a more generic
method for adding another column of data?
On Tue, 2005-27-09 at 10:39 -0400, tom wright wrote:
Tom
# 
Got it thanks
the trick was to create a variable with the new name
str_idx<-paste('c',i,sep='')
then use double square brackets
l_cooef[str_idx]]<-whatever()


f_haardisolve<-function(v_dataset){
    #pad data to make length a power of 2
    v_dataset<-f_paddata(v_dataset)

    l_cooef<-list() #holder for cooefficents
    i_count<-1      #identity counter

    while(length(v_dataset)>0){
        #seperate odd and even points
        v_dataset<-f_splitpoints(v_dataset)

        v_dataset<-f_haarpredict(v_dataset)
        v_dataset<-f_haaruplift(v_dataset)
        
        str_idx<-paste('c',i_count,sep='')
        l_cooef[[str_idx]]<-list()

l_cooef[[str_idx]]<-v_dataset[c((length(v_dataset)/2)+1:(length(v_dataset)/2))]
        v_dataset<-v_dataset[c(1:length(v_dataset)/2)]

        i_count<-i_count+1
    }
    l_cooef
}
On Tue, 2005-27-09 at 10:39 -0400, tom wright wrote:
Vincent Goulet
# 
Le 27 Septembre 2005 10:39, tom wright a ??crit??:
That's an exercise I give to my students! This will create the sequences:
You can then assign the names with
but 'data$1' will not work. You will need either 'data$"1"' or 'data[["1"]]'. 
I'd use different names...

HTH.
Gabor Grothendieck
# 
When sapply is used on a character vector it will use those
as the names so:

data <- sapply(as.character(1:5), function(x) seq(as.numeric(x),10))

will give data with the required names.
On 9/27/05, tom wright <tom at maladmin.com> wrote: