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
-
Swap the first and last columns
22579 Solvers
-
1220 Solvers
-
Getting the indices from a vector
11836 Solvers
-
How many rectangles in a grid ?
78 Solvers
-
135 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.