Vector.random uses the wrong distribution

The (https://github.com/mateogianolio/vectorious/wiki/Vector-API#random)[docs] state that Vector.random samples based on a normal distribution using the passed mean and standard deviation. What the code actually does is using a uniform distribution, which is not even centered around the passed mean, but around (mean+deviation)/2.

Here is the code in vector.js:

Vector.random = function (count, deviation, mean, type) {
    deviation = deviation || 1;
    mean = mean || 0;
    return Vector.fill(count, function() {
      return deviation * Math.random() + mean;
    }, type);
  };

(A detail: when 0 is passed as the deviation, it should not be reset to 1.)

[eweitnauer]

The pull request was only addressing an incorrect falsy test of the function arguments. The issue I described is still the same: Math.random() picks a number from a uniform distribution in [0,1). The code assumes that it returns a normal distribution centered around 0.

[mateogianolio]

https://arxiv.org/pdf/1303.6257.pdf