manuelgloeckler
manuelgloeckler
Hey @WangYuxiang8, thanks, for reporting this! Sorry, something we forgot to test :/ I will have a look to fix this. This function should make the objects compatible with the...
I encountered and fixed this in the current draft for #979 (while trying to get all tests running for the new interface). The problem is that the `conditional_potential` does not...
Great, thanks for the review and suggested changes :) I think it actually also would be more appropriate to adapt the naming to: - `Estimator` -> `ConditionalEstimator` - `DensityEstimator` ->...
Great :). I just checked with @michaeldeistler, who also works on the Density estimator functionality in #1066; so, we will delay merging this until #1066 is done.
This currently won't work due to #1154.
Okey SNPE stuff works now. MCMC stuff still needs to be done and tested.
Okay, it should be done now. Rebasing on main for some reason did recommit everything ... . A small summary: - One can improve `accept_reject`. I have already tracked all...
Oh wow thanks for fixing this mess :D. I am not sure what went wrong on the rebase (but probably merging main in should be preferred if there were many...
@janfb I have already responded to the reshape (but you can only see it in code review ???, and I cannot answer the new comment you made). In short, This...
Okay, the SNPE_A test fails because the SNPE_A posterior switches the "batching" behavior in different rounds - After first round > Sample from the proposal are of shape (500, 1,...