How can I automatically scale each of the elements of a Vector V to the respective size of the cell of the struct S?

I have a struct called Observations.t0 which has 140 cells -> Observations(1).t0 up to Observations(140).to and every Observation has a different amount of elements.

I want to create a Vector with ones with the length of the cell for every cell and multiply it with a value from another struct. I tried it this way:

Ratings=vertcat(ones(length(Observations(:).t0),1)*values.b(:,4));

If Observations(1).t0 had 5 elements in its cell, I want to create a vector that looks like this (1;1;1;1;1) and afterwards multiply the entire vector with the value of values.b(1,4)) which could be 3. The result would be the vector (3;3;3;3;3). Afterwards I want to do that process again with the second cell and extend my vector with the results. That could look like this: (3;3;3;3;4;4).

I get an error with the length(Observations(:).t0),1) it says to many input arguments. Can I solve that with a for loop?

Answer 1

[Wilson A N]

You can use for loop to do this as shown in the code below:

   clear;
   clc;

   structSize = 20;
   observations(structSize).t0 = [];
   for i = 1:structSize
       observations(i).t0 = ones(1,i);
   end

   values.b = rand(20,20);

   Ratings = [];

   for i = 1:4
       Ratings = [Ratings;ones(length(observations(i).t0),1)*values.b(i,4)];
   end

[Stephen23]

Note that expanding the array Ratings on each loop iteration is not efficient and will show an mlint warning:

To write efficient code it is important to preallocate the array before the loop:

https://www.mathworks.com/help/matlab/matlab_prog/preallocating-arrays.html

Answer 2

[Stephen23]

You can use arrayfun to do this quite neatly:

>> vec = [1,9,4,0];
>> obs(1).t0 = 0:4;
>> obs(2).t0 = 5;
>> obs(3).t0 = 6:7;
>> obs(4).t0 = 8:9;
>> C = arrayfun(@(s,n)n*ones(size(s.t0)),obs,vec,'uni',0);
>> out = horzcat(C{:})
out =
     1     1     1     1     1     9     4     4     0     0