Split vector in MATLAB

Question

I\'m trying to elegantly split a vector. For example,

vec = [1 2 3 4 5 6 7 8 9 10]

According to another vector of 0\'s and 1\'s of the same

Answer 1

You can do this with a combination of find and arrayfun:

vec = [1 2 3 4 5 6 7 8 9 10];
N = numel(vec);
cut = [0 0 0 1 0 0 0 0 1 0];
ind = find(cut);
ind_before = [ind-1 N]; ind_before(ind_before < 1) = 1;
ind_after = [1 ind+1]; ind_after(ind_after > N) = N;
out = arrayfun(@(x,y) vec(x:y), ind_after, ind_before, 'uni', 0);

We thus get:

>> celldisp(out)

out{1} =

     1     2     3         

out{2} =

     5     6     7     8    

out{3} =

    10

---

So how does this work? Well, the first line defines your input vector, the second line finds how many elements are in this vector and the third line denotes your cut vector which defines where we need to cut in our vector. Next, we use find to determine the locations that are non-zero in cut which correspond to the split points in the vector. If you notice, the split points determine where we need to stop collecting elements and begin collecting elements.

However, we need to account for the beginning of the vector as well as the end. ind_after tells us the locations of where we need to start collecting values and ind_before tells us the locations of where we need to stop collecting values. To calculate these starting and ending positions, you simply take the result of find and add and subtract 1 respectively.

Each corresponding position in ind_after and ind_before tell us where we need to start and stop collecting values together. In order to accommodate for the beginning of the vector, ind_after needs to have the index of 1 inserted at the beginning because index 1 is where we should start collecting values at the beginning. Similarly, N needs to be inserted at the end of ind_before because this is where we need to stop collecting values at the end of the array.

Now for ind_after and ind_before, there is a degenerate case where the cut point may be at the end or beginning of the vector. If this is the case, then subtracting or adding by 1 will generate a start and stopping position that's out of bounds. We check for this in the 4th and 5th line of code and simply set these to 1 or N depending on whether we're at the beginning or end of the array.

The last line of code uses arrayfun and iterates through each pair of ind_after and ind_before to slice into our vector. Each result is placed into a cell array, and our output follows.

---

We can check for the degenerate case by placing a 1 at the beginning and end of cut and some values in between:

vec = [1 2 3 4 5 6 7 8 9 10];
cut = [1 0 0 1 0 0 0 1 0 1];

Using this example and the above code, we get:

>> celldisp(out)

out{1} =

     1

out{2} =

     2     3         

out{3} =

     5     6     7

out{4} =

     9         

out{5} =

    10