Seth Axen

@matbesancon @mschauer @andreasnoack could any of you review this PR? In particular, note the discussion in https://github.com/JuliaStats/Distributions.jl/pull/1422#discussion_r749140387 as to the suitability of this contribution to Distributions.

On second look, the current version does not even support one of the allowed signatures, namely, when `rng` is a vector of `AbstractRNG`'s, as `rand(rng)` would then select a random...

Yeah, that's type piracy and unnecessarily allocates. Just moving the call to `rand` into the broadcast does what you want. e.g. ```julia julia> using Random julia> x, y = randn(10),...

Although perhaps I'm misunderstanding what a vector of RNGs means here. I had interpreted it as a separate RNG per chain, but with `rand_coupled` it seems the intention is for...

Perhaps the only potentially complicated part of this is that the transition has a jittered integration time, where the jitter could be with an RNG or a pre-determined Halton sequence...

See also #70

As noted by @nsiccha on Slack, the reference code for the paper is available at https://github.com/stan-dev/stan/compare/develop...bbbales2:stan:experimental/warmup-wishart. A good first step would be to implement a low-rank update Euclidean metric type....

> As noted by @nsiccha on Slack, the reference code for the paper is available at [stan-dev/[email protected]:stan:experimental/warmup-wishart](https://github.com/stan-dev/stan/compare/develop...bbbales2:stan:experimental/warmup-wishart). Specifically, these files seem to contain all code for the metric adaptation, while...

Python 3 does, yes, but Python 2 doesn't. And while many popular Python packages have dropped Python 2 support, it's harder for others because systems like CentOS, which is widely...

Thanks for the reference! I think after #228 and #214 are merged, it will be worth revisiting this. In addition to making code more efficient, adopting a mutating pattern should...