Repeating elements (vector)

Hi everyone,
Suppose I have matrix A
A = [1; 2; 3; 4]
And I want to have matrix B
B = [1; 1; 2; 1; 2; 3; 1; 2; 3; 4]
I don't want to use loop to do so because I have lots of observations. So how can I do this efficiently without using loop? Thank you very much!
Jeff

5 Comments

@Tsz Heung: what is the rule to generate B from A ?
What do you mean? My question is how can I generate B from A.
I assume the rule is you want just the 1st row, then the first 2 rows, then the first 3 rows, etc, all collapsed into a single vector.
@Tsz Heung: There is an infinite number of methods to create B based on A from your example. To create a program, we have to know the definition of the procedure. One example is not enough to define this unequivocally.
@Adam: You mean:
  • one time the first row
  • one time the first 2 rows
  • two times the first 3 rows
  • one time the last row
???
Adam
Adam on 19 Jan 2017
Edited: Adam on 19 Jan 2017
No, I meant
  • one time the first row,
  • one time the first 2 rows,
  • one time the first 3 rows,
  • one time the first 4 rows,...

Sign in to comment.

 Accepted Answer

Stephen23
Stephen23 on 19 Jan 2017
Edited: Stephen23 on 19 Jan 2017
There are most likely much more efficient algorithms than this:
>> N = 4;
>> X = (1:N)'*ones(1,N);
>> X(triu(true(N)))
ans =
1
1
2
1
2
3
1
2
3
4
You can use X as indices into your vector to select the values that you want.

2 Comments

+1: An excellent guess.
Really Appreciated! Thank you Stephen!

Sign in to comment.

More Answers (1)

nonzeros(triu(repmat(A,1,length(A))))
Note: this assumes that none of the entries are 0.

2 Comments

Stephen23
Stephen23 on 19 Jan 2017
Edited: Stephen23 on 19 Jan 2017
+1 very nice. It would be interesting to know if repmat or multiply is faster:
nonzeros(triu((1:N)'*ones(1,N)))
Really Appreciated! Thank you Walter!

Sign in to comment.

Categories

Asked:

on 19 Jan 2017

Commented:

on 19 Jan 2017

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!