Closed spar7453 closed 4 years ago
Hello,
Uniform(T, T).sample(n)
returns T
types vector which has length n
.
So, if you want to generate float uniform random array then use Uniform(0f64, 1f64)
instead.
(Documentation is not updated. I'll update documentation asap.)
But for Normal
, we don't expect integer or unsigned integer samples. Thus, for Normal
only float type samples are allowed.
thank you
I am not sure whether it is intended or not but
Uniform(0,1) has type TPDist and
Uniform(0,1).sample(10) type generates random integers between [0, 1), so only zeros
However, Normal(0,1) has the same type TPDist and
Nomral(0,1).sample(10) genrates random floats from the distribution