Adding valies from a table into a zero matrix

Hello,

I've created a zero matrix array of the size that I need, and I want to add values from a table to it.

My table is in an excel spreadsheet with three columns. Column 1 denotes the x co-ordinate, column 2 the y co-ordinate, and column 3 the value I want to add. I'm trying to create a loop that will allow me to find the matrix position (x, y) as specified in the table, and then add the third column value to that position only.

I know roughly how to specify positions in each, but not how to then say, 'use the numbers in here to find the position in the array'.

If my matrix is X and table is A, I'm thinking the code needs to be similar to:

i=1

b=A[i,1]

c=A[i,2]

d=A[i,3]

X[b,c] + d

i=i+1

Any help on how to smooth this out would be fantastic as it's been forever since I did anything with tables and matrixes other than pulling values out.

Cheers.

[Azzi Abdelmalek]

Make your question clear and brief. Have you a problem to read your Excel file? If not, how your are data are stored? like this ?

M=[1 2 3;
   4 5 6;
   7 8 9]

Then What is the expected result from M?

[Maeve]

My excel file has read in fine an example is:

M=[8 26 240; 39 6 52; 600 530 211]

I want to use this excel file to find X(8,26) and add 240 to that 0, then find X(39,6) and add 52 to that zero.

But to use a loop to specify the x coordinates based off of the M table as I have hundreds of these co-ordinates to find.

Answer 1

[Azzi Abdelmalek]

M=[8 26 240; 39 6 52; 600 530 211]
idx=ismember(M(:,1:2),[8,26],'rows')
out=M(idx,3)

[Maeve]

Thank you, that's helpful but I don't think it's going to solve my problem. I'm looking at how to use the table to change my zero array.

What I want to do is take

X=[0 0 0; 0 0 0; 0 0 0]

Use

A=[1 2 12; 1 3 8; 2 2 5; 3 3 16]

To turn X into

out=[0 12 8; 0 5 0; 0 0 16].

[Azzi Abdelmalek]

Ok, but you need to explain how to get out from X and A

[Maeve]

That is what I need help with, as I said in my original question.

A specifies co-ordinates in X and a value with which to replace that zero, which will then give me the modified X Out.

[Azzi Abdelmalek]

We can't provide the criterion at your place. From the example, explain how to get the result, then we can help you to code it

[Maeve]

If I have A [1 2 12] (1,2) refers to a co-ordinate that I want to find in X, 12 refers to the number I want to add to the zero at X(1,2).

I don't know any other way to explain what I'm looking to do.

[Azzi Abdelmalek]

X=zeros(3)
A=[1 2 12; 1 3 8; 2 2 5; 3 3 16]
idx=sub2ind(size(X),A(:,1),A(:,2))
X(idx)=A(:,3)

[Maeve]

That looks like it could work, cheers! I'll give it a try.

Answer 2

[Guillaume]

A loop is totally overkill for this. There are two simple ways of obtaining your final output:

1. sub2dind which will convert two vectors of row, column coordinates into a vector of linear indices. You then use this linear index to directly assign all the values in one go:

X = zeros(3, 3);
X(sub2ind(size(X), A(:, 1), A(:, 2)) = A(:, 3);

2. Create a sparse matrix with your vectors of coordinates and vector of values and convert to full matrix:

%no need for zero
X = full(sparse(A(:, 1), A(:, 2), A(:, 3), 3, 3));