How can I choose non-repeating values with a probability of 90% and 10% ?

Hello friends, how can I choose values ​​that do not repeat with a probability of 90% and 10%?

let me explain myself better.

For example I have a vector:

vector = [43 19 63 60 19 27 6]

So what I have done so far with my algorithm is that if the numbers that are inside the vector are not repeated, I assign a binary code of "1", on the other hand, if any number inside the vector is repeated, I assign a binary code of "0". ", as can be seen in the image below.

now I want to solve the following problem, that, if the numbers within the vector are not repeated, they have a 90% probability of assigning the binary code "1" and a 10% probability of assigning the binary code "0", on the other hand if the numbers within the vector are repeated with a 100% probability of being assigned the binary code "0".

for example:

taking the variable :

vector = [43 19 63 60 19 27 6]

numbers that are not repeated we have 43,63,60,27 and 6.

then it would have to be randomly assigned the binary code "1" with a 90% probability and the binary code "0" with a 10% probability.

taking into account that the probability is assigned randomly

then the result would be something like this.

I touch 90% to 43 so it is assigned the binary code "1"

I touch 90% to 63 so it is assigned the binary code "1"

I touch 10% to 60 so it is assigned the binary code "0"

I touch 90% to 27 so it is assigned the binary code "1"

I touch 90% to 6 so it is assigned the binary code "1"

Now the repeated numbers of the vector are 19,19.

then with 100% the binary code of "0" is assigned

19 is assigned the binary code "0" >> with 100%

19 is assigned the binary code "0" >> with 100%

in the end my matrix would look like this

43 "1"

19 "0"

63 "1"

60 "0"

19 "0"

27 "1"

6 "1"

thank you very much in advance

[Wasseem Khardawi]

if you share the code, maybe i can help

Answer 1

[Voss]

To generate a number that has value 1 with 90% probability and value 0 with 10% probability, you can use randi() to generate a random integer between 1 and 10, then use the condition that the integer is greater than 1. It will be greater than 1 90% of the time (so the condition will be true) and (less than or) equal to 1 10% of the time (and the condition will be false). Convert the condition from logical to double, and you have your number(s). A demonstration with 10000 random numbers obeying this distribution:

x = double(randi(10,1,10000) > 1);
nnz(x)

ans = 9000