Given n input vectors x1, x2, …, xn, generate a p*n matrix y whose rows contain all element-wise combinations of the vectors x1, x2, …, xn, where p = numel(x1)*numel(x2)*…* numel(xn). The rows of y are organized in an "increasing" ordering; see examples below.
- Example 1 (character array, n = 2): Input: x1 = 'ab'; x2 = 'cde'; Output:
y = ['ac'
'ad'
'ae'
'bc'
'bd'
'be'] - Example 2 (numeric array, n = 3): Input: x1 = [0 1], x2 = [0 1]; x3 = [0 1]; Output:
y = [0 0 0
0 0 1
0 1 0
0 1 1
1 0 0
1 0 1
1 1 0
1 1 1]You may assume that all inputs are valid vectors whose lengths are at least 1.
Note: The Statistics and Machine Learning toolbox provides two useful functions for full factorial design, namely ff2n (for binary case) and fullfact (for general cases). However, they do not directly accomplish the task in Example 1. The purpose of this problem is to create a toolbox independent function to achieve our goal. Have fun!
Solution Stats
Problem Comments
Solution Comments
Show commentsProblem Recent Solvers28
Suggested Problems
-
2369 Solvers
-
Replace NaNs with the number that appears to its left in the row.
3068 Solvers
-
middleAsColumn: Return all but first and last element as a column vector
651 Solvers
-
Create an n-by-n null matrix and fill with ones certain positions
725 Solvers
-
193 Solvers
More from this Author28
Problem Tags
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!
It would be better if the test suite did not rely on a particular ordering of the rows. You could use isempty(setxor(myfullfact(...), y_correct, 'rows')) instead of isequal(...)
Thanks for your suggestion. But I prefer an "increasing" ordering of the rows because that could be more useful in some applications. I have made it clear in the problem description that ordering is required.