Rémi Louf
Rémi Louf
If we note $m$ the scale parameter and $b$ the shape parameter, my understanding is that JAX [implemnts](https://jax.readthedocs.io/en/latest/_modules/jax/_src/scipy/stats/pareto.html#logpdf): $$P(x) = \frac{b m^b}{(x-loc)^{b+1}}$$ While on the other hand PyMC3 implements $$P(x)...
`shape` and `scale` make more sense to me, and it is better to have an API close to PyMC3's. You can keep `loc`.
Hi @tblazina Looks like the tests are not passing :( Are you still planning on working on this?
Hi @elanmart, Thank you for taking the time to share this with me! Regarding what you identified: 1. Indeed, I just noticed that recently. It is indeed problematic if you...
That's strange regarding `A-B`, I identified the problem 10 days ago and I thought I'd fixed it. Are you running the latest version (latest commit)? Unfortunately no workaround for the...
> I'm really sorry for generating noise :cry: No worries, you're really helpful :) > Do you want me to close this ticket and open a clean one for `rng_key`...
> some of which I suppose are (or will soon be invalid) invalid after the `
Just to let you know, I'll make some time to work on this and the other issue one my NUTS PR is merged on BlackJAX (which means MCX will support...
Great! If you remember which ones don't hesitate to open issues now.
> I thought I would open an issue to follow up on [#1 (comment)](https://github.com/rlouf/mcx/issues/1#issuecomment-812733048) , even though I'm not sure if I would be able to contribute a PR. Yes...