How to convert 4D double matrix into 2D arrays

Hi

I'm sorry for my English :(

I'm trying to convert 4-D double matrix into 2D arrays

There are a total of 14 different scenarios, each represented by a matrix of size 1x101 in the program. The representation in 4D is structured as (:,:,1,1) through (:,:,101,1) for the first scenario, (:,:,1,2) through (:,:,101,2) for the second scenario, and so on, up to (:,:,1,14) through (:,:,101,14) for the fourteenth scenario. Each set of indices within the 4D structure corresponds to a unique value in the respective scenario.

Thank you so much

[Rik]

How exactly to you want to put that into 2D? Or do you perhaps mean you want to loop through all 1414 combinations and do some calculation with the 2D part?

[Sung Wook Choi]

I want a 101x1 matrix for each of the 14 scenarios, where there are 14 different radii, with a sigma(1) for each wavelength (101).

[Dyuman Joshi]

"There are a total of 14 different scenarios, each represented by a matrix of size 1x101 in the program."

Then there should be a 3D array, 1x101x14 or 14x1x101. Where did the 4th dimension come from?

Is it a singleton dimension? In that case, what is the size of the 4D array?

"I want a 101x1 matrix for each of the 14 scenarios"

Why not store it as 101x14 matrix and use indexing to access the data?

Could you attach the data for the 4D array?

[Sung Wook Choi]

Yes, I believe it's a singleton dimension. I'm also curious about why it's portrayed as a 4D double array. How can I verify its size? It's currently labeled as a 4D double array.

[Dyuman Joshi]

"I'm also curious about why it's portrayed as a 4D double array."

Most likely, because that's how it was defined.

"How can I verify its size?"

Use size

[Sung Wook Choi]

Do you mean 'using size' to verity its size?

I used sz=size( ), and it says

sz=1 1 101 14

[Dyuman Joshi]

Yes, that's what I meant.

Ok, there are 2 singleton dimensions.

Now, What is the output you want? A 101x14 array?

[Sung Wook Choi]

Yes I want it!

Answer 1

[Dyuman Joshi]

Use squeeze -

in = rand(1,1,101,14);
size(in)

ans = 1×4

1 1 101 14

out = squeeze(in);
size(out)

ans = 1×2

101 14

[Sung Wook Choi]

Thank you so much! However, I have one question. After creating a matrix of size 101 x 14 using commands, do I need to manually copy/paste and assign values for each element? Or can I input the command directly into the existing matrix, and will it automatically reshape into a 101 x 14 format?

And when I try to simply copy/paste a 4D double matrix, there is a following message. What does it mean to use a cell array instead?

"You cannot paste data from the clipboard into an array. Note that MATLAB numeric arrays cannot contain text. If you want to paste text into an array, consider using a cell array instead."

[Dyuman Joshi]

"Or can I input the command directly into the existing matrix"

What is the command that you are trying to use?

"And when I try to simply copy/paste a 4D double matrix"

From where are you trying to copy the data?

[Sung Wook Choi]

1) I meant the command is sigma.

I want to focus on sigma(#1) of each wavelength(#101) for different 14 radiis(#14)

2) It's also the sigma that I want to copy.

ex) in sigma(4-D double), val(:,:,1,1) = 5.6851e-15 / val(:,:,2,1)= 5.6722e-15 ,,,,

the matrix is written in this way, and numerical values represent the desired sigma for me.

[Dyuman Joshi]

I don't think I understood what you said.

Could you attach the data? Use the paperclip button to attach.

[Sung Wook Choi]

sorry for my expression ,,

Here it is! I want to deal with 'sigma' inside the file 'scat_amor_sweep'

[Dyuman Joshi]

I hope this is clear now -

load('scat.mat')
%Check the contents of the array
scat_amor_sweep

scat_amor_sweep = struct with fields:
    Lumerical_dataset: [1×1 struct]
                    f: [101×1 double]
               lambda: [101×1 double]
               radius: [3.5000e-08 4.0000e-08 4.5000e-08 5.0000e-08 5.5000e-08 6.0000e-08 6.5000e-08 7.0000e-08 7.5000e-08 8.0000e-08 8.5000e-08 9.0000e-08 9.5000e-08 1.0500e-07]
                sigma: [1×1×101×14 double]

%Assign the data to a variable
in = scat_amor_sweep.sigma;
size(in)

ans = 1×4

1 1 101 14

%Remove the singleton dimensions from the variable
out1 = squeeze(in)

out1 = 101×14

1.0e-12 * 0.0057 0.0084 0.0119 0.0163 0.0217 0.0283 0.0352 0.0424 0.0495 0.0561 0.0627 0.0692 0.0753 0.0880 0.0057 0.0083 0.0117 0.0159 0.0212 0.0277 0.0347 0.0419 0.0491 0.0558 0.0625 0.0690 0.0753 0.0881 0.0057 0.0082 0.0116 0.0156 0.0208 0.0271 0.0341 0.0414 0.0487 0.0555 0.0622 0.0689 0.0751 0.0882 0.0057 0.0082 0.0115 0.0154 0.0204 0.0266 0.0336 0.0410 0.0483 0.0551 0.0620 0.0687 0.0750 0.0882 0.0057 0.0081 0.0113 0.0151 0.0199 0.0261 0.0330 0.0404 0.0479 0.0548 0.0617 0.0684 0.0749 0.0882 0.0057 0.0081 0.0112 0.0149 0.0195 0.0255 0.0324 0.0399 0.0474 0.0545 0.0614 0.0682 0.0747 0.0882 0.0058 0.0081 0.0111 0.0147 0.0192 0.0250 0.0319 0.0394 0.0470 0.0541 0.0611 0.0680 0.0745 0.0882 0.0059 0.0081 0.0110 0.0145 0.0188 0.0245 0.0313 0.0389 0.0465 0.0537 0.0608 0.0677 0.0743 0.0880 0.0060 0.0081 0.0110 0.0143 0.0185 0.0240 0.0307 0.0383 0.0461 0.0534 0.0605 0.0674 0.0740 0.0879 0.0061 0.0082 0.0109 0.0142 0.0182 0.0235 0.0300 0.0377 0.0456 0.0531 0.0602 0.0672 0.0738 0.0877

You can combine the above steps as well -

out2 = squeeze(scat_amor_sweep.sigma)

out2 = 101×14

1.0e-12 * 0.0057 0.0084 0.0119 0.0163 0.0217 0.0283 0.0352 0.0424 0.0495 0.0561 0.0627 0.0692 0.0753 0.0880 0.0057 0.0083 0.0117 0.0159 0.0212 0.0277 0.0347 0.0419 0.0491 0.0558 0.0625 0.0690 0.0753 0.0881 0.0057 0.0082 0.0116 0.0156 0.0208 0.0271 0.0341 0.0414 0.0487 0.0555 0.0622 0.0689 0.0751 0.0882 0.0057 0.0082 0.0115 0.0154 0.0204 0.0266 0.0336 0.0410 0.0483 0.0551 0.0620 0.0687 0.0750 0.0882 0.0057 0.0081 0.0113 0.0151 0.0199 0.0261 0.0330 0.0404 0.0479 0.0548 0.0617 0.0684 0.0749 0.0882 0.0057 0.0081 0.0112 0.0149 0.0195 0.0255 0.0324 0.0399 0.0474 0.0545 0.0614 0.0682 0.0747 0.0882 0.0058 0.0081 0.0111 0.0147 0.0192 0.0250 0.0319 0.0394 0.0470 0.0541 0.0611 0.0680 0.0745 0.0882 0.0059 0.0081 0.0110 0.0145 0.0188 0.0245 0.0313 0.0389 0.0465 0.0537 0.0608 0.0677 0.0743 0.0880 0.0060 0.0081 0.0110 0.0143 0.0185 0.0240 0.0307 0.0383 0.0461 0.0534 0.0605 0.0674 0.0740 0.0879 0.0061 0.0082 0.0109 0.0142 0.0182 0.0235 0.0300 0.0377 0.0456 0.0531 0.0602 0.0672 0.0738 0.0877

[Sung Wook Choi]

I really appreciate your help! Thanks to you, I've achieved the desired results. Additionally, do you happen to know a more efficient way to convert such a matrix to a text file? Currently, I'm copying/pasting the matrix from 'out2' to Excel and then copying/pasting it again into a text file. I'm curious if there's a faster and more efficient method.

[Dyuman Joshi]

Use writematrix

[Sung Wook Choi]

Thank you so much! Finally, may I ask what the difference is between the 'out1' and 'out2' matrices? They appear to be the same, and I couldn't identify any distinctions.

Also, when you use 'squeeze,' it removes two singletons. Does 'squeeze' mean removing all singletons in a matrix?

[Dyuman Joshi]

You are welcome!

Yes, they are the same.

For the second one, I have just combined all the steps in one command.

Yes, 'squeeze' removes all the singleton dimensions of an array.