Jan

I suggest, one the other PR #1188 is merged, we can merge `main` into this branch to then double check whether CPU saturation has improved with the fixes. If yes,...

My two cents: 1. Good point by @gmoss13 , I remember we had a z-scoring bug in the `zuko` flow builders, see https://github.com/sbi-dev/sbi/pull/1492. This is not released yet, so maybe...

Yes, sounds good! In general, @satwiksps , Kudos for offering your help with many different issues here, that's great! Your plans and PRs are very structured and well-reviewable - much...

> Sorry, only diving into this now: I think even more useful than a `max_sampling_time` would be a flag `reject_outside_prior_support: bool`. Using this flag, the user could then manually check...

Yes, on a closer look, SNPE-C needs the snapshots of the neural net weights from previous rounds when evaluating the mog-based proposal in non-atomic SNPE-C, e.g., here: https://github.com/sbi-dev/sbi/blob/ebcd68e0c9a6772626d625a8cbfe6fcffa3f2820/sbi/inference/trainers/npe/npe_c.py#L178 and then...

@Dev-Sudarshan are you able to reproduce the CI error locally when you install things from scratch? e.g., in a fresh `uv` environment: - `uv venv -p python3.12` - `source .venv/bin/activate`...

I see, thanks for checking @Dev-Sudarshan ! So, we added pymc>=5.0.0,

I gave it a go with ChatGTP Codex and here's the result: ## Problem - Traced the crash to PyMC’s new Rich-based progress bar: the column "{task.fields[step_size]:0.2f}" now receives a...

Great that you tackle this @Dev-Sudarshan ! I agree with @JuliaLinhart , we can just focus on MLP-based C2ST for the GPU support for L-C2ST. The different C2ST options are...

I suggest we find an implementation that uses the sbi settings by default, and when a user passes additional options, those settings will be updated for the `kwargs` the user...