Split vector by parts with more evenly distributed length

While my previous question got answered fine (and I did not specify it good enough), I'm looking for a way to split a vector so that the sizes of the subvectors are more evenly distributed. Let me explain with the following example.

So again let's say I have the following vector

V = 1:2184
P = [floor(length(V/127)) & ceil(length(V/127))] %[17 18]

Now I want to split the vector somewhat more distributed, so:

          sv1 sv2 sv3 sv4 sv5  ... sv123 sv124 sv125 sv126 sv127        %subvector;
  Vbad =  [17  17  17  17  17  ...  17    20]                           %sv length; don't want this
  Vgood = [17  17  17  17  17  ...  17    17    18    18     18]        %sv length; looking for this; always end up with 127 subvectors

Thanks in advance.

[Ameer Hamza]

It appears that you are not even running the code before posting here. Have you computed the output of the following line

P = [floor(length(V/127)) & ceil(length(V/127))] %[17 18]

what you actually meant is

 P = [floor(length(V)/127)];

Answer 1

[Ameer Hamza]

As @dbp pointed out, the problem still isn't well defined, but based on your example the answer will only contain P or P+1 elements.

V = 1:2184;
numberPartition = 127;

P = floor(length(V)/numberPartition);
L = length(V) - numberPartition*P;
partitionElements = [P*ones(1, numberPartition-L) (P+1)*ones(1, L)];

partitions = mat2cell(V, 1, partitionElements);

[YT]

I agree that I wasn't able to define the problem better (lack the specific terms and vocabulary to do so), but this is what I was looking for.

Answer 2

[dpb]

Consider the following...

>> for k=fix(sqrt(L)):-1:2,if rem(L,k)==0,disp([L,k,L/k]),end,end
      2184          42          52
      2184          39          56
      2184          28          78
      2184          26          84
      2184          24          91
      2184          21         104
      2184          14         156
      2184          13         168
      2184          12         182
      2184           8         273
      2184           7         312
      2184           6         364
      2184           4         546
      2184           3         728
      2184           2        1092
>>

gives you every factor evenly divisible into your vector length; one of these choices (and only one of these) will be the value that is exactly even. If you have some other criteria on how many divisions or a target size, then you may have to heuristically pick another value.

It still isn't well-enough defined to know what are hard requirements vis a vis desires.