How to find the minimum difference between the 3 elements of a vector in app designer?

I need to group the elements of a matrix 3 by 3 with the minimum difference between them. I found something like that: min(min(abs(X(1)-X2))), but l have 3 values.

[Dyuman Joshi]

Let's say the input is this -

y = magic(3)

y = 3×3

8 1 6 3 5 7 4 9 2

What should be the output? and what logic is followed to obtain the output?

[piston_pim_offset]

I have a matrix like:

X = rand(5,1)

X = 5×1

0.6076 0.2072 0.0161 0.2327 0.9077

The output should be the three of them with minumum difference among them, [0.0161 0.2327 0.2072]

[Dyuman Joshi]

What is the minimum difference here? How is it calculated?

[piston_pim_offset]

This is actually what l'm asking for. Output is three of the elements of X with minimum difference among them.

[Dyuman Joshi]

I am asking what logic did you follow to get that output from that particular input.

How do you define minimum difference?

[piston_pim_offset]

It was an axample. l didn't use anything, just looked at the numbers and wrote it.

Answer 1

[DGM]

I take this to mean the three elements of X with the smallest extreme spread. I don't see that pdist2() would work for that.

This might work, so long as the number of elements in X stays small.

X = randi([0 99],3,3)

X = 3×3

65 46 98 71 61 28 91 4 26

C = nchoosek(X(:),3);
Crange = range(C,2);
row = Crange == min(Crange);
C(row,:)

ans = 1×3

65 71 61

From what's given, there's no reason to think that there's a unique minimum combination. The first example @Dyuman Joshi gave is a good demonstration. The example I gave will return all combinations with the same minimum spread.

[piston_pim_offset]

Thanks @DGM, it worked exactly. Can we make it to give all the elements in the matrix grouped in threes?

[DGM]

That's what C is. C is the set of possible combinations of the elements of X, if taken 3 at a time. Obviously, the size of C will increase rapidly as the number of elements of X increases.

[piston_pim_offset]

l mean if we have 12 elements in the matrix X, can we make the code to give 4 groups from X which provide the requirement?

[DGM]

You mean combinations of length 4?

X = randi([0 99],3,4) % 12 elements

X = 3×4

33 42 21 23 34 45 21 38 66 15 86 10

C = nchoosek(X(:),4); % all combinations of length 4
Crange = range(C,2);
row = Crange == min(Crange);
C(row,:)

ans = 1×4

15 21 21 23

[piston_pim_offset]

No, l meant grouping all elments 3 by 3, so length is the same as 3. But it gives other combinations giving minimum distance between them. In the code above, it gives an only C array, but there rest of the matrix X stays there untouched. I wanted to group all elements till there is none of them left.

Actually, tried and made it by deleting the elements in X linked with C array, then l'm writing this combinations to a table.

It would be grate if you have any other idea.

Thank you for your responses @DGM!

[DGM]

I still don't understand. How does a 3x3 array have 12 elements, and where does 4 come in?

Are you trying to find the minimal combinations left in X after removing each minimal combination?

[piston_pim_offset]

-Are you trying to find the minimal combinations left in X after removing each minimal combination?

+Absolutely!

Answer 2

[Sulaymon Eshkabilov]

Try to implement pdist2()

[piston_pim_offset]

l'll try.