Ok so I am clustering data into clusters which are then indexed using a column. The data is in the form of motion vectors and so my data will look like this after being clustered:
[index x y x' y']
for example:
[1 3 5 4 6;
1 4 6 5 7;
2 3 5 4 6;
2 8 9 9 3;
3 2 3 2 4]
in above array there are 3 clusters, with clusters 1 and 2 each containing 2 vectors.
My problem is that I sometimes have to delete clusters based on certain criteria, and may be left with:
[2 3 5 4 6;
2 8 9 9 3;
3 2 3 2 4]
I want to be able to correct the index after deletion, so that it starts at 1 and ends with the number of clusters. So in this case replace the 2s with 1s and 3s with 2s.
Im sure there must be a simple way using a for loop but Ive been trying for a while and can't get ti right?
Assuming your matrix is called data, try this:
>> data = [2 3 5 4 6;
2 8 9 9 3;
3 2 3 2 4]
data =
2 3 5 4 6
2 8 9 9 3
3 2 3 2 4
>> data(:,1) = cumsum(diff(data([1 1:end], 1)) ~= 0) + 1
data =
1 3 5 4 6
1 8 9 9 3
2 2 3 2 4
A simple call to unique will help you do that. You can use the third output of it to assign each unique and new ID using the first column of the new data matrix (index vector) to replace its first column. Also, make sure you use the 'stable' flag so that it assigns IDs in order of occurrence from top to bottom:
%// Data setup
A = [1 3 5 4 6;
1 4 6 5 7;
2 3 5 4 6;
2 8 9 9 3;
3 2 3 2 4];
%-----
B = A(3:end,:); %// Remove first two rows
%// Go through the other IDs and reassign to unique IDs from 1 up to whatever
%// is left
[~,~,id] = unique(B(:,1), 'stable');
%// Replace the first column of the new matrix with the new IDs
B(:,1) = id; %// Replace first column with new IDs
We get:
>> B
B =
1 3 5 4 6
1 8 9 9 3
2 2 3 2 4
Related
As the title says, I want to find all rows in a Matlab matrix that in certain columns the values in the row are equal with the values in the previous row, or in general, equal in some row in the matrix. For example I have a matrix
1 2 3 4
1 2 8 10
4 5 7 9
2 3 6 4
1 2 4 7
and I want to find the following rows:
1 2 3 4
1 2 3 10
1 2 4 7
How do I do something like that and how do I do it generally for all the possible pairs in columns 1 and 2, and have equal values in previous rows, that exist in the matrix?
Here's a start to see if we're headed in the right direction:
>> M = [1 2 3 4;
1 2 8 10;
4 5 7 9;
2 3 6 4;
1 2 4 7];
>> N = M; %// copy M into a new matrix so we can modify it
>> idx = ismember(N(:,1:2), N(1,1:2), 'rows')
idx =
1
1
0
0
1
>> N(idx, :)
ans =
1 2 3 4
1 2 8 10
1 2 4 7
Then you can remove those rows from the original matrix and repeat.
>> N = N(~idx,:)
N =
4 5 7 9
2 3 6 4
this will give you the results
data1 =[1 2 3 4
1 2 8 10
4 5 7 9
2 3 6 4
1 2 4 7];
data2 = [1 2 3 4
1 2 3 10
1 2 4 7];
[exists,position] = ismember(data1,data2, 'rows')
where the exists vector tells you wheter the row is on the other matrix and position gives you the position...
a less elegant and simpler version would be
array_data1 = reshape (data1',[],1);
array_data2 = reshape (data2',[],1);
matchmatrix = zeros(size(data2,1),size(data1,1));
for irow1 = 1: size(data2,1)
for irow2 = 1: size(data1,1)
matchmatrix(irow1,irow2) = min(data2(irow1,:) == data1(irow2,:))~= 0;
end
end
the matchmatrix is to read as a connectivity matrix where value of 1 indicates which row of data1 matches with which row of data2
Given the matrix I = [1,2;3,4], I would like to duplicate the elements to create a matrix I2 such that:
I2 = [1 1 1 2 2 2
1 1 1 2 2 2
1 1 1 2 2 2
3 3 3 4 4 4
3 3 3 4 4 4
3 3 3 4 4 4]
Other than using repmat, what other methods or functions are available?
Use kron:
>> N = 3 %// Number of times to replicate a number in each dimension
>> I = [1,2;3,4];
>> kron(I, ones(N))
ans =
1 1 1 2 2 2
1 1 1 2 2 2
1 1 1 2 2 2
3 3 3 4 4 4
3 3 3 4 4 4
3 3 3 4 4 4
This probably deserves some explanation in case you're not aware of what kron does. kron stands for the Kronecker Tensor Product. kron between two matrices A of size m x n and B of size p x q creates an output matrix of size mp x nq such that:
Therefore, for each coefficient in A, we take this value, multiply it with every value in the matrix B and we position these matrices in the same order as we see in A. As such, if we let A = I, and B be the 3 x 3 matrix full of ones, you thus get the above result.
Using indexing:
I = [1, 2; 3, 4]; %// original matrix
n = 3; %// repetition factor
I2 = I(ceil(1/n:1/n:size(I,1)), ceil(1/n:1/n:size(I,2))); %// result
One-liner with bsxfun -
R = 3; %// Number of replications
I2 = reshape(bsxfun(#plus,permute(I,[3 1 4 2]),zeros(R,1,R)),R*size(I,1),[])
Sample run -
I =
3 2 5
9 8 9
I2 =
3 3 3 2 2 2 5 5 5
3 3 3 2 2 2 5 5 5
3 3 3 2 2 2 5 5 5
9 9 9 8 8 8 9 9 9
9 9 9 8 8 8 9 9 9
9 9 9 8 8 8 9 9 9
I have two matrix A and B. Suppose I would like to find in each row of matrix A the smallest number, and for the same cell that this number is in Matrix A, do find the corresponding number of the same cell in matrix B. For example the number in matrix A will be in the position A(1,3), A(2,9)...and I want the corresponding number in B(1,3), B(2,9)... Is it possible to do it, or I am asking something hard for matlab. Hope someone will help me.
What you can do is use min and find the minimum across all of the rows for each column. You would actually use the second output in order to find the location of each column per row that you want to find. Once you locate these, simply use sub2ind to access the corresponding values in B. As such, try something like this:
[~,ind] = min(A,[],2);
val = B(sub2ind(size(A), (1:size(A,1)).', ind));
val would contain the output values in the matrix B which correspond to the same positions as the minimum values of each row in A. This is also assuming that A and B are the same size. As an illustration, here's an example. Let's set A and B to be a random 4 x 4 array of integers each.
rng(123);
A = randi(10, 4, 4)
B = randi(10, 4, 4)
A =
7 8 5 5
3 5 4 1
3 10 4 4
6 7 8 8
B =
2 7 8 3
2 9 4 7
6 8 4 1
6 7 3 5
By running the first line of code, we get this:
[~,ind] = min(A,[],2)
ind =
3
4
1
1
This tells us that the minimum value of the first row is the third column, the minimum value of the next row is the 4th column, and so on and so forth. Once we have these column numbers, let's access what the corresponding values are in B, so we would want row and columns (1,3), (2,4), etc. Therefore, after running the second statement, we get:
val = B(sub2ind(size(A), (1:size(A,1)).', ind))
val =
8
7
6
6
If you quickly double check the accessed positions in B in comparison to A, we have found exactly those spots in B that correspond to A.
A = randi(9,[5 5]);
B = randi(9,[5 5]);
[C,I] = min(A');
B.*(A == repmat(C',1,size(A,2)))
example,
A =
2 1 6 9 1
2 4 4 4 2
5 6 5 5 5
9 3 9 3 6
4 5 6 8 3
B =
3 5 6 8 1
9 2 9 7 1
5 6 6 5 6
4 6 1 4 5
5 3 7 1 9
ans =
0 5 0 0 1
9 0 0 0 1
5 0 6 5 6
0 6 0 4 0
0 0 0 0 9
You can use it like,
B(A == repmat(C',1,5))
ans =
9
5
5
6
6
5
4
1
1
6
9
This question already has answers here:
Generate all possible combinations of the elements of some vectors (Cartesian product)
(4 answers)
Closed 8 years ago.
So I'm writing a program to determine the unique combinations of a beaded necklace, but I can't seem to get it right. The rules are you can't have the same necklace forwards and backwards, and you can't have the same necklace with one bead being slid around to the other end. I've attached some pictures to clarify.
I wrote the code for it, and I thought I had achieved what I was trying to do, but it's not working correctly.
n = [1 2 3 4 2 4];
% green = 1
% blue = 2
% yellow = 3
% red = 4
p = perms(n);
total = max(size(p));
for i = 1:max(size(p))
q = p;
q(i) = [];
for j = 1:max(size(q))
if isequal(p(i),fliplr(q(j)))
total = total - 1;
elseif isequal(p(i),circshift(q(j),[1,1]))
total = total - 1;
elseif isequal(p(i),circshift(q(j),[length(q(j))-1,length(q(j))-1]))
total = total - 1;
end
disp(total)
end
end
Logically, this makes sense to me, but I could just be crazy.
If the problem size is small, you can vectorize all the comparisons (using bsxfun):
n = [1 2 3 4 2 4];
%// green = 1
%// blue = 2
%// yellow = 3
%// red = 4
N = numel(n);
p = perms(n).'; %'// generate all permutations
p2 = NaN([size(p) N+1]); %// this will store permutations with flips and shifts
p2(:,:,1) = p; %// original
p2(:,:,2) = flipud(p); %// flips
for k = 1:N-1
p2(:,:,2+k) = circshift(p,k); %// circular shifts
end
eqElem = bsxfun(#eq, p, permute(p2, [1 4 2 3]));
eqMat = squeeze(any(all(eqElem, 1), 4)); %// 1 if equal
remove = any(tril(eqMat, -1), 1); %// remove permutations that are "similar"
%// to a previous one, where "similar" means "equal up to circular shifts or
%// flips"
result = p(:,~remove).'; %'// all valid arrangements; one per row
resultNum = size(result, 1); %// number of arrangements
Results:
result =
1 3 2 2 4 4
1 3 2 4 4 2
1 3 2 4 2 4
1 3 4 2 2 4
1 3 4 2 4 2
1 3 4 4 2 2
1 2 3 2 4 4
1 2 3 4 2 4
1 2 3 4 4 2
1 2 2 3 4 4
1 2 2 4 4 3
1 2 2 4 3 4
1 2 4 3 2 4
1 2 4 3 4 2
1 2 4 2 3 4
1 2 4 2 4 3
1 2 4 4 2 3
1 2 4 4 3 2
1 4 4 3 2 2
1 4 4 2 2 3
1 4 4 2 3 2
1 4 3 4 2 2
1 4 3 2 2 4
1 4 3 2 4 2
1 4 2 3 2 4
1 4 2 3 4 2
1 4 2 2 3 4
1 4 2 2 4 3
1 4 2 4 2 3
1 4 2 4 3 2
resultNum =
30
You should do p = unique(p,'rows') before any loops. To see why, call perms([1 1 1]) at the command line.
There are a few issues here:
1) p, the perms, is a 2D matrix, so to get each perm you need to do p(i,:) to get the row. p(i) is just a single number.
2) You don't remove wrong answers from your list, so you will check against them twice. For example, say the first in the list is [1 2 3 4 2 4]; and the second is [4 2 4 3 2 1];. The fliplr check will compare these two combinations twice, once in the first loop around, once in the second.
3) If you want to make sure that any permutation which is a rotation is excluded (not just moving one bead around), you'll need some more circshift.
Consider using ismember with rows option again to compare a single row (e.g. a flipped version of the row you're checking) to an entire matrix.
I am trying to flip certain parts of a matrix. I can explain better by example. Let's say that I have a matrix
M = [ 1 3 6;
1 2 4;
1 7 1;
2 9 0;
2 8 3;
2 4 2;
2 3 1;
3 6 5;
3 4 5;
3 1 9;
4 2 4;
4 8 6 ]
What I'd like to do here is take any rows with an even number in the first column, and flip the third column elements. The end result would look like this:
1 3 6
1 2 4
1 7 1
2 9 1 *
2 8 2 *
2 4 3 *
2 3 0 *
3 6 5
3 4 5
3 1 9
4 2 6 *
4 8 4 *
Note the rows marked with a star have had the elements of the third column flipped upside-down. The problem I'm having is going through each row like in a for-loop you cannot flip an entire set of rows.
Thanks in advance for any help.
Another time accumarray is the way to go:
A =[ 1 3 6 ;
1 2 4 ;
1 7 1 ;
2 9 0 ;
2 8 3 ;
2 4 2 ;
2 3 1 ;
3 6 5 ;
3 4 5 ;
3 1 9 ;
4 2 4 ;
4 8 6 ]
C = accumarray(A(:,1),A(:,3),[],#(x) {flipud(x)} ); %// get groups according to
%// first column and flip it
C = vertcat(C{:}); %// cell array returned,
%// transform to matrix
mask = ~mod(A(:,1),2); %// mask for even numbers
A(mask,3) = C(mask); %// replace masked values of 3rd column with flipped ones
returns:
A =
1 3 6
1 2 4
1 7 1
2 9 1
2 8 2
2 4 3
2 3 0
3 6 5
3 4 5
3 1 9
4 2 6
4 8 4
Certainly slower, but just for fun in two lines:
C = accumarray(A(:,1),A(:,3),[],#(x) {flipud(x)} );
A(~mod(A(:,1),2),3) = getfield( vertcat(C{:}), {~mod(A(:,1),2)});
%// well no, I won't explain it...
Edit: I assumed your first column just contains integers!
I would suggest you break the problem down into stages, something like so:
Identify blocks you wish to flip
Extract them
Flip them
Replace them
You can identify a set of even numbers using the unique and mod functions, then use a for loop over them and use logical indexing to pull/replace the blocks.
Here, try this
a = magic(5); % Some data in a 5x5 matrix
b = 1:numel(a); % Indices of <a>
Rearrange b however you want, then do a=a(b) to reassign a based on the reassigned indices of b. For example, the following code
disp(a(b));
would just return the elements of a in their original order. For your application this code should work:
a = <your matrix data>
b = 1:numel(a);
b = [b(1:27) fliplr(b(28:31)) b(32:34) fliplr(b(35:36))] % Change this part
a = reshape(a(b),size(a))
You should change b based on whatever you need it to do.