Giuseppe Carleo

yes this would be useful

I guess I might have mixed up the threads here, but indeed looks like @femtobit is also suggesting to add (back) the method to compute the derivatives of the log...

maybe there is no good default behavior here for the sampler, I guess it's best just to ask it always be specified?

Yes, this would be great! On Sun, Nov 28, 2021, 11:13 Filippo Vicentini ***@***.***> wrote: > #1001 Made me realise that > if somebody writes a model that outputs the...

Wouldn't it be better to use mypy instead? On Thu, Apr 8, 2021, 19:03 Vladimir Vargas-Calderón < ***@***.***> wrote: > In another issue > I > proposed the use of...

I would target full compatibility my mypy since (hopefully) we might even able to get some speed up using mypyc if types are correctly annotated

yes I agree... the only issue is that in principle it would still make sense to do ```log_overlap(x)``` ... :) x is just a very special ```VariationalState```. Maybe (not sure)...

indeed ```log_amplitude``` is specific for pure states ... but ```log_value``` is a bit too generic, I am not sure... let's do ```log_value``` for the moment and then we can just...

Yes please, that would be helpful! On Sun, Mar 31, 2019, 14:57 Anil Radhakrishnan wrote: > Hi, I recently came across this repository and it seems very useful. > I...

what are the different curves?