splitting a vector into separate vectors

Hi Matlab community,

I have a vector like

V = [2 2 2 2 2 4 4 4 7 7 8 9]

I want to separate this vector into

V1 = [2 2 2 2 2 ], V2=[4 4 4 ], V3=[7 7 ], V4=8, V5=9

Suppose that I do not know the value in vector V. In each iteration, I have a new vector.

Thanks

Answer 1

[Stephen23]

A much better approach using a cell array:

V = [2,2,2,2,2,4,4,4,7,7,8,9]

V = 1×12

2 2 2 2 2 4 4 4 7 7 8 9

<mw-icon class=""></mw-icon>

<mw-icon class=""></mw-icon>

L = diff(find([1,diff(V),1]))

L = 1×5

5 3 2 1 1

<mw-icon class=""></mw-icon>

<mw-icon class=""></mw-icon>

C = mat2cell(V,1,L)

C = 1x5 cell array
    {[2 2 2 2 2]}    {[4 4 4]}    {[7 7]}    {[8]}    {[9]}

[Neda]

Thank you so much, Stephen.

Now, how can I access the resutl? I mean how can I understand how many elements are there in the first array? and second array?....

I mean, how many 2 we have here?

{[2 2 2 2 2]}

[Umeshraja]

Hi @Neda,

You can use indexing to extract the individual arrays and then use functions like 'length' or 'numel' to determine the number of elements in each array.

firstArray = C{1};
numElementsInFirstArray = numel(firstArray);

Please refer to the following documentation to know more

https://www.mathworks.com/help/matlab/ref/double.numel.html

[Stephen23]

"Now, how can I access the resutl? I mean how can I understand how many elements are there in the first array? and second array? I mean, how many 2 we have here?"

My answer already tells you that, take a look at the variable L, it gives you the lengths of each vector.

You can access the arrays using very basic MATLAB indexing, eg. the 2nd vector:

C{2} % the vector
L(2) % its length

You do not need to call an extra function like LENGTH or NUMEL, because the information is already there.

Answer 2

[Steven Lord]

Can you dynamically create variables with numbered names like V1, V2, V3, etc.? Yes.

Should you do this? The general consensus is no. That Discussions post explains why this is generally discouraged and offers several alternative approaches.

[Neda]

Thank you Steven,

Actually, my problem is not the name, splitting the matrix is my problem.

[Neda]

Thank you so much Steven. Very helpful !!!

Answer 3

[Umeshraja]

Hi @Neda,

If you are working with large 1D vectors. using a binary search might be a good option to optimize time complexity.

Below is the sample code:

function separatedVectors = separateUsingBinarySearch(V)
    % Initialize an empty cell array to store the separated vectors
    separatedVectors = {};
    n = length(V);   
    
    if n == 0
        return;
    end
    
    startIdx = 1; 
    
    % Continue until the start index exceeds the length of the vector
    while startIdx <= n
        value = V(startIdx);  
        endIdx = findEndIndex(V, startIdx, n, value);
        % Store the current group of identical elements in the cell array
        separatedVectors{end+1} = V(startIdx:endIdx);
        startIdx = endIdx + 1;
    end
end

function endIdx = findEndIndex(V, startIdx, n, value)
    low = startIdx; 
    high = n; 
    
    % Perform binary search to find the last occurrence of 'value'
    while low < high
        mid = floor((low + high + 1) / 2); 
        if V(mid) == value
            low = mid; % Move the lower bound up if the mid value matches
        else
            high = mid - 1; % Move the upper bound down if the mid value does not match
        end
    end
    
    endIdx = low; 
end

% Example usage
V = [2 2 2 2 2 4 4 4 7 7 8 9 9 9 9]; 
separatedVectors = separateUsingBinarySearch(V);

% Display the results
for i = 1:length(separatedVectors)
    fprintf('[%s]\n', num2str(separatedVectors{i}));
end

[2  2  2  2  2]
[4  4  4]
[7  7]
[8]
[9  9  9  9]

The function separateUsingBinarySearch efficiently divides a sorted vector V into sub-vectors, each containing consecutive identical elements. For each starting index, findEndIndex uses binary search to find the last occurrence of the current value efficiently. The sub-vector from startIdx to endIdx is extracted and added to separatedVectors. The process repeats by updating startIdx to just after endIdx for the next group.