How can I convert a 3D matrix to 2D matrix in the way specified in this post

Hello, I am new to Matlab and am having a bit of trouble shaping a 3D matrix in a certain way.

For example, I would like to convert 3D matrix M:

 val(:,:,1) =    
1    1    1    1    1
1    1    1    1    1
1    1    1    1    1
1    1    1    1    1
val(:,:,2) =    
2    2    2    2    2
2    2    2    2    2
2    2    2    2    2
2    2    2    2    2
val(:,:,3) =    
3    3    3    3    3
3    3    3    3    3
3    3    3    3    3
3    3    3    3    3

into the following 2D matrix:

1    1    1    1    1
1    1    1    1    1
1    1    1    1    1
1    1    1    1    1 
2    2    2    2    2
2    2    2    2    2
2    2    2    2    2
2    2    2    2    2 
3    3    3    3    3
3    3    3    3    3
3    3    3    3    3
3    3    3    3    3

Using reshape I can convert the matrix into a 2D form, however the end result (which I don't want) is similar to this:

1    2    3    4    5
1    2    3    4    5
1    2    3    4    5
1    2    3    4    5 
1    2    3    4    5
1    2    3    4    5
1    2    3    4    5
1    2    3    4    5 
1    2    3    4    5
1    2    3    4    5
1    2    3    4    5
1    2    3    4    5

What is an efficient way to achieve the desired 2D matrix orientation?

EDIT:

While the example above shows val(:,:,1) through val(:,:,3), the script is meant to allow from val(:,:,1) through val(:,:,n) where n can be any number I need it to be.

Additionally, the dimensions of each "val" are 4x5 in this example, but could be any dimension, according to variables specified.

Essentially, I would like an (a,b,c) matrix to become an (a,b) matrix, where the (a,b) sections are stacked in order from 1 to c.

Thank you, and sorry for the lack of clarity in the original question.

Answer 1

[Roger Stafford]

   result = reshape(permute(val,[1,3,2]),[],size(val,2)); % <-- Corrected

[Neuro Guy]

I'm sorry, I should have clarified. I will edit the original question to reflect this:

While the example above shows val(:,:,1) through val(:,:,3), the script is meant to allow from val(:,:,1) through val(:,:,n) where n can be any number I need it to be.

Additionally, the dimensions of each "val" are 4x5 in this example, but could be any dimension, according to variables specified.

Essentially, I would like an (a,b,c) matrix to become an (a,b) matrix, where the (a,b) sections are stacked in order from 1 to c.

[Roger Stafford]

My code is supposed to be general enough to encompass the changes you mention. Try it and see.

[Neuro Guy]

This worked perfectly!

Thank you.

[Akash kumar]

1.) If you want to add matrix in Vertical direction [Roger stafford:- I support it.]

result = reshape(permute(val,[1,3,2]),[],size(val,2));

2.) If you want to add matrix in Horizontal direction then

M=reshape(A,size_of_the_row,[]) % But size of the each row is equal.

for example:- According to the "Neuro Guy" first post:- each submatrix row size=4

then, M=reshape(A,4,[]); then output is (According to the 2.) point)

1 1 1 1 1 2 2 2 2 2 3 3 3 3 3

1 1 1 1 1 2 2 2 2 2 3 3 3 3 3

1 1 1 1 1 2 2 2 2 2 3 3 3 3 3

1 1 1 1 1 2 2 2 2 2 3 3 3 3 3

(According to the 1.) point):- Output is

1 1 1 1 1

1 1 1 1 1

1 1 1 1 1

1 1 1 1 1

2 2 2 2 2

2 2 2 2 2

2 2 2 2 2

2 2 2 2 2

3 3 3 3 3

3 3 3 3 3

3 3 3 3 3

3 3 3 3 3

Answer 2

[Star Strider]

This works:

val = cat(3, ones(4,5), ones(4,5)*2, ones(4,5)*3);
val2D = reshape(val, 5, [])'

val2D =

       1     1     1     1     1
       1     1     1     1     1
       1     1     1     1     1
       1     1     1     1     1
       2     2     2     2     2
       2     2     2     2     2
       2     2     2     2     2
       2     2     2     2     2
       3     3     3     3     3
       3     3     3     3     3
       3     3     3     3     3
       3     3     3     3     3

[Neuro Guy]

I'm sorry, I should have clarified. I will edit the original question to reflect this:

While the example above shows val(:,:,1) through val(:,:,3), the script is meant to allow from val(:,:,1) through val(:,:,n) where n can be any number I need it to be.

Additionally, the dimensions of each "val" are 4x5 in this example, but could be any dimension, according to variables specified.

Essentially, I would like an (a,b,c) matrix to become an (a,b) matrix, where the (a,b) sections are stacked in order from 1 to c.

[Star Strider]

In your more general description, my code is essentially:

val2D = reshape(val, b, [])'

That will work to convert any of your 3D matrices to a 2D matrix of the sort you describe. The ‘[]’ is a sort of ‘wildcard’ that will be calculated to produce the correct output.

[Neuro Guy]

Ah, so that's what that is.

Thank you for clarifying that.

[Star Strider]

My pleasure.

I should have been more explicit. To make my code completely general, it becomes:

val2D = reshape(val, size(val,2), [])'

and to make an anonymous function out of it that you can put at the start of your code and then use anywhere:

Mtx3to2 = @(M) reshape(M, size(M,2), [])';

calling it as:

val2D = Mtx3to2(val);

You can of course name it whatever you want. You don’t have to call it ‘Mtx3to2’. It’s just that I’m tired just now and out of more creative ideas for function names.

[Neuro Guy]

This will work great. Thank you for the wisdom.

[Star Strider]

My pleasure.

I would have preferred that you Accept my Answer.

Answer 3

[Azzi Abdelmalek]

a=permute(val,[2 1 3])
out=a(:,:)'

[Neuro Guy]

I'm sorry, I should have clarified. I will edit the original question to reflect this:

While the example above shows val(:,:,1) through val(:,:,3), the script is meant to allow from val(:,:,1) through val(:,:,n) where n can be any number I need it to be.

Additionally, the dimensions of each "val" are 4x5 in this example, but could be any dimension, according to variables specified.

Essentially, I would like an (a,b,c) matrix to become an (a,b) matrix, where the (a,b) sections are stacked in order from 1 to c.