how to fulfill the matrix with for loop?

Hello there,

xxx=[1 3 8 20 30 40 50];
inn=[0.1 0.3 0.7 0.2 0.4 0.6 0.5];
for i=1:(length(xxx)-1)
    l=xxx(i);
    u=xxx(i+1);
    ve(:,l:u)=inn(i);
    i=i+1;
end

I want to create a vector which has 0.1 from 1 to 3, 0.3 from 4 to 8, 0.7 from 9 to 20 and so on. However, the code gives 0.1 from 1 to 2, 0.3 from 3 to 7 and so on. i=i+1 does not work. How can i make it correct?

I have a second question.

In my original code, xxx=[0 3 8 20 30 40 50]; it starts with zero, Matlab says that "Subscript indices must either be real positive integers or logicals." since it reuires to start from 1. But according to my aim, it must start from 0. Could you also help me to correct it, please?

Thanks in advance.

Answer 1

[Walter Roberson]

xxx=[0 3 8 20 30 40 50];
inn=[0.1 0.3 0.7 0.2 0.4 0.6 0.5];
count = diff(xxx);
ve = repelem(inn(1:length(count)), count);
ve

ve = 1×50

0.1000 0.1000 0.1000 0.3000 0.3000 0.3000 0.3000 0.3000 0.7000 0.7000 0.7000 0.7000 0.7000 0.7000 0.7000 0.7000 0.7000 0.7000 0.7000 0.7000 0.2000 0.2000 0.2000 0.2000 0.2000 0.2000 0.2000 0.2000 0.2000 0.2000

[Paul Kaufmann]

TIL: repelem exists! Very nice, and also, obviously much cleaner than my approach.

[esra kan]

Thank you, Walter!

Answer 2

[Paul Kaufmann]

the intended dimension of ve is not quite clear to me, but maybe this is what you want:

x = [0 3 8 20 30 40 50]
n = [ 0.1 0.3 0.7 0.2 0.4 0.6 0.5]
arb = 3; % arbitrary dimension, up to you
dx = diff(x);
v = [];
for i = 1:numel(dx)
    v = [v ; ones(dx(i),arb)*n(i)]
end

This method above is computationally very inefficient, but it gets the job done, if your matrix is relatively small.

This is the output result:

>> v
v =
    0.1000    0.1000    0.1000
    0.1000    0.1000    0.1000
    0.1000    0.1000    0.1000
    0.3000    0.3000    0.3000
    0.3000    0.3000    0.3000
    0.3000    0.3000    0.3000
    0.3000    0.3000    0.3000
    ...       ...       ...
>> whos v
  Name       Size            Bytes  Class    
  v         50x3              1200  double   

[esra kan]

Hello Paul, thanks for your reply. I need 1x50. Also, this is just a sample, actually my data is very large.