Re: [Numpy-discussion] python is cool

Roland Schulz wrote:

Hi,

I think the best way to solve this issue to not use a state at all. It is fast, reproducible even in parallel (if wanted), and doesn't suffer from the shared issue. Would be nice if numpy provided such a stateless RNG as implemented in Random123: www.deshawresearch.com/resources_random123.html

Roland

That is interesting. I think np.random needs to be refactored, so it can accept a pluggable rng - then we could switch the underlying rng.