Rémi Louf

What it is not about: - Warmup using Adam instead of Dual Averaging - Dynamical HMC - [Discontinuous HMC](https://arxiv.org/abs/1705.08510) - [Riemannian manifold Hamiltonian Monte Carlo](https://arxiv.org/abs/1212.4693) - Metropolis-within-Gibbs - Particle Filtering...

I took it off because the implementation would be a lot different; the design I had in my was that when we use `.do(a=10)` the variable `a` now becomes an...

Now that we're moving samplers to [blackjax](https://github.com/blackjax-devs/blackjax), I think tests based on SBC would be better there. It will also be a lot simpler to implement than in MCX.

Just to say I haven't forgotten about this, I just need some time to think about it. I use commit hooks locally too :) All I can say for now...

I have tried it in the `blackjax` repo, and I think it would be a useful addition to `mcx`.

## Let the user know what is happening A user should get at least some feedback, at best interesting information, when performing an action. - [x] Progress bars are a...

A thought for people using `mcx` in a production environment: the sampler should provide clear and actionable logs.

There definitely is an interest in implementing as many distributions as is possible! I would first rank roughly by how often you think they're used, if that's not the case...

I would open a PR in JAX.

Sure! Here's a tentative list of distributions that we still have to implement: https://github.com/rlouf/mcx/issues/74