Formulation to Matlab code

How to write this formulation in Matlab:

is it :

for i=1:N
    S(i) = -PT(i) * log(PT(i));
end

[Dimitris Kalogiros]

What is the base of the logarithm? Is it 10 ? If so, then you must use log10() .

[Maroco Sc]

I have checked several papers, they did not mention the base of logarithm. if it is 10, then the code will be:

for i=1:N
    S(i) = -PT(i) * log10(PT(i));
end

Right?

Answer 1

[Steven Lord]

There's no need to loop. The log and log10 functions can operate on arrays of data, and as long as you use element-wise multiplication you can do this in one line. [Actually, I'll need two; one to create sample data and one for the actual operation.]

PT = 10*rand(10);
S = -PT.*log10(PT);

You can compare this with the result of doing the same operation one element at a time to see if they're different.

S2 = zeros(size(PT));
for whichelement = 1:numel(PT)
    S2(whichelement) = -PT(whichelement).*log10(PT(whichelement));
end

S-S2

Answer 2

[Raj]

Since your equation asks for 'log' not 'ln' , I think the code should be:

for i=1:N
    S(i) = -PT(i) * log10(PT(i));
end