How to rearrange values in a matrix

Hello, I have a large matrix with 3 columns(x,y,and z) and I want to rearrange the values inside this matrix. Example:

*Initial matrix

x y z

1 1 5

1 2 6

1 3 7

2 1 8

2 2 9

2 3 2

*Final matrix My goal is to rearrange the values of the matrix in this form

y
x 1 2 3
1 5 6 7 <-z values
2 8 9 2

Answer 1

[Sean de Wolski]

There are a few elegant ways to do this. I would use accumarray

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

V = accumarray(data(:,[1 2]),data(:,3))

Now V does not have the index values, but you already know them:

rr = 1:size(V,1)  %row
cc = 1:size(V,2)  %column

So I would just keep them separate.

[afrya]

Thanks for your answer, could you tell me an another way to rearraenge the values and keeping the index values inside the matrix?

[Sean de Wolski]

See Andrei's answer below.

Or add to mine:

V = [[nan cc];[rr(:),V]]]

[afrya]

Thanks

[afrya]

When I'm working with decimal numbers,I have the following error

x =

    0.6000    0.6000    5.0000
    0.6000    0.8000    6.0000
    0.6000    0.9000    7.0000
    0.8000    0.6000    8.0000
    0.8000    0.8000    9.0000
    0.8000    0.9000    2.0000

>> V = accumarray(x(:,[1 2]),x(:,3)) Error using accumarray First input SUBS must contain positive integer subscripts.

Do you know how to solve this problem?

[Andrei Bobrov]

see my answer

Answer 2

[Andrei Bobrov]

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

out = nan(max(xyz(:,1:2))+1);
out(2:end,1) = unique(xyz(:,1));
out(1,2:end) = unique(xyz(:,2));
out(2:end,2:end) = accumarray(xyz(:,1:2),xyz(:,3));

ADD

x =[    0.6000    0.6000    5.0000
0.6000    0.8000    6.0000
0.6000    0.9000    7.0000
0.8000    0.6000    8.0000
0.8000    0.8000    9.0000
0.8000    0.9000    2.0000];

[r,ii,ii] = unique(x(:,1));
[c,jj,jj] = unique(x(:,2));
v = accumarray([ii,jj],x(:,3));
out = [nan,c';r,v];

[afrya]

Thanks for your answer

[afrya]

Hello,

I'm working with a large matrix that contains a lot of zero values.When I transform this matrix from a list table to a cross table, I have more zero values in it.I want to keep all the initial zeros values in it and supress all the zero values added from matlab for the calculation of the average.Here is an example of what I want to do:

x=

    0.6000    0.6000    5.0000
    0.6000    0.8000    6.0000
    0.6000    0.8500         0
    0.6000    0.8200         0
    0.8000    0.8000    9.0000
    0.8000    0.9000    2.0000
List table to cross table

       NaN    0.6000    0.8000    0.8200    0.8500    0.9000
    0.6000    5.0000    6.0000         0         0         0   Average1
    0.8000         0    9.0000         0         0    2.0000   Average2

Average1=(5+6+0+0)/4=2,75

Average2=(9+2)/2=5,5 Thanks in advance

[Andrei Bobrov]

x=[ 0.6000    0.6000    5.0000
    0.6000    0.8000    6.0000
    0.6000    0.8500         0
    0.6000    0.8200         0
    0.8000    0.9000    2.0000
    0.8000    0.8000    9.0000];

[rc,~,c1] = arrayfun(@(ii)unique(x(:,ii)),1:2,'un',0);
v = accumarray([c1{:}],x(:,3));
out = [nan,rc{2}',nan;rc{1},v,sum(v,2)./sum(v>0,2)];

or other last row

out = [nan,rc{2}',nan;rc{1},v,sum(v,2)./numel(rc{2})];

[afrya]

Thank you for your answer,it helped me. Now, what I want to do is to calculate the average for each row but without taking into account the zeros added by matlab.

1st row: there is only one zero added by matlab for this coordinate(0.6;0.9) So , i would like to calculate this average: Average1=(5+6+0+0)/4=2.75

2nd row: there is 3 zeros added by matlab for this coordinates(0.8;0.6),(0.8;0.82),(0.8;0.85) so i'd like to calculate this average, Average2=(9+2)/2=5.5

I have no idea how to do that knowing that i'm working with a large matrix, thanks in advance

[Andrei Bobrov]

another variant of last row

out = [nan,rc{2}',nan;rc{1},v,accumarray(c1{1},x(:,3),[],@mean)];

[afrya]

Thanks for your help, it works now