根据每个矩阵上函数的结果从列表中删除单个矩阵-在r中

Question

I have a list of matrices - list1. These matrices are named x1, x2, x3..... x999, x1000. I have a custom function (fun1) that I run like this:

result<-sapply(list1,fun1)

This function will then return the individual results of this function like this (Example):

x1    x2    x3    x4    x5....   x999  x1000
3     2     0     0     2 ....    9    0

What I would like to do is remove from the list all those that return a result of '0' from fun1. So above, I would remove (presumably among others in the list) x3, x4, x1000. Ideally, the new list created (list2) following the removal of matrices that return 0s from fun1 would keep the original names (x1, x2 etc.).

I have struggled with working out how to do this. I have thought about trying to make a function that would be applied to the 'result' object that uses ifelse to see if a 0 is contained, before then applying another function that removes the matrix from list1:

sapply(result, fun.remove)


fun.remove <- function(x){
              ifelse (result = 0,   
              function.delete.matrix,
              NA))
              }

I'm not sure what should go in the function.delete.matrix, or if using ifelse is even the best way to go about it.

Thanks for any suggestions.

Answer 1

You can do this with simple subsetting

list2 <- list1[result!=0]

Here's some sample data:

set.seed(15)
list1<-as.list(rpois(20,1)); 
names(list1)<-paste0("x", seq_along(list1))

and let

fun1<-function(x) x;
result<-sapply(list1,fun1)

so

list1[result==0]

shows all the elements with 0 values

and

list1[result!=0]

shows those that do not have 0s.