Jan

icon indicating a website link janfb.github.io

icon company @aai-institute icon location Munich icon location Researcher @aai-institute. Working on Bayesian Machine Learning and Simulation Based Inference

Results 241 comments of Jan

sampling via pymc raises TypeError since pymc v5.20.1

I see, thanks for checking! and pinning the version to `"pymc>=5.0.0,

sampling via pymc raises TypeError since pymc v5.20.1

Good, then we have a solution for now. The tests already have a conditional flag for the 5.21 version, right? We can just keep that. So, the only change would...

tests: switch to a bi-modal posterior for testing

Thanks for tackling this @etouron1 the simulator and resulting bi-modal posterior you suggested makes sense 👍 I assume the test is run with the vanilla NPE? Yes, it looks reasonable...

bug: docs building is failing because of missing score posterior

Yes, that's it. yes, would be great if you can fix it for `mkdocs` @StarostinV . (We will have to fix it for Readthedocs as well @michaeldeistler , but this...

bug: docs building is failing because of missing score posterior

solved with #1641

docs: move general class docstrings from __init__ to class level for better API refs

Cool! I see, in [this commit ](https://github.com/sbi-dev/sbi/pull/1519/commits/8fff0fecc50b8a647ea2ecdc33955532650206bd) there is now one sentence in each docstring. I am wondering whether we should add more content? that content now usually lives in...

docs: a how-to-guide for using unconditional flows to detect model misspecification

and there is a how-to-guide here: https://sbi.readthedocs.io/en/latest/how_to_guide/18_model_misspecification.html

bug: pymc nuts failing with ValueError

this seems to be a duplicate of #1397

docs: add a tutorial / how-to-guide for using pyro with sbi

> Is there a place where it would make sense to also document use of a learned posterior as a prior in a Pyro model? > > The killer app...

improve interface for custom density estimators passed to inference classes

fixed with #1633