Adrien Corenflos

This should do: the broadcasting should be smart enough to handle everything. ```python import particles.distributions as dists import numpy as np import scipy.stats as stats def log_det_chol(chol): diags = np.diagonal(chol,...

Hi @jeanfeydy Thanks for the quick response. Maybe I should have been a bit more specific :) I am interested in the gradient of the coupling matrix w.r.t. the inputs,...

Hey, I had the same issue when playing with the library for early tests of one of my papers. The regularisation param does not have the same meaning in pot...

I had the same issue, the scaling you're using is too close to 1. @jeanfeydy I fixed it locally by using a generator. Do you want me to send a...

Jean will correct me if I'm mistaken, but the real fix for you is to lower your scaling parameter I'd presume, except if you know you really, really can't do...

Opened PR https://github.com/blackjax-devs/blackjax/pull/117

Opened issue https://github.com/blackjax-devs/blackjax/issues/118

Ready for review. For some reason the precommit thingy does not correct the notebook. I am not sure where exactly there is a pb with it...

> For some reason the precommit thingy does not correct the notebook. I am not sure where exactly there is a pb with it... Essentially it says it fixed it...

Also made residual resampling allow for sub and supersampling while at it.