Benjamin Kurt Miller

Maybe we need an object that lives between store and dataset? I think it's likely that accessing zarr with `Dataset` is just too slow.

This is an important issue, but it will have to be dealt with in a future release. I appreciate all the work you've done it already @fnattino and everyone.

For inspiration, consider the [dynesty implementation of Priors](https://github.com/joshspeagle/dynesty/blob/9fc19cfeec17ce3d87ba16c962e6ca59cd21d548/priors.py). We need to be able to collect several distributions together for a prior, including dependent ones... but still have access to cdf,...

One option is to require that new distributions subclass pytorch distribution, but that seems a bit excessive to me.

@rogerkuou is this solved?

Nice find! I'll take a look.

This would be nice for the next release, for reproducibility when dealing with multiprocessing.

The suggestion on that site is to leave the master with executed notebooks and have a dev branch without. imo swyft should download instantly and right now it takes some...

I encountered the saving and loading standards of `__getstate__ ` and `__setstate__`. Perhaps we can use those going forward? https://realpython.com/lessons/getstate-setstate/

> We are interested by this functionality. @iandewancker what is the status on your side, do you have some spare time to work on this ? > > @AshBT note...