Martin Modrák

Not 100% sure what your aim is - do you need to do SBC for such model (and thus a workaround would be enough) or do you want to implement...

Thanks for reporting this. The underlying problem here is that R does not distinguish between a scalar and vector of length 1 while C++ and Stan do. In best case...

Note to self: the currently developed system for variable attributes could be used to resolve this.

The current implementation of brms backends relies only on `rename_pars` - then it uses the inner `fit$fit` object to get draws. So as long as `rename_pars` works for the converted...

Hi, thanks for reporting. You cannot force the simulator to cover all of the values everytime. My impression is that you would like to have a prior on the parameters...

I agree that more modularization would be sensible and playing more nicely with target is definitely something we'd like to have. The good news is that I think that most...

Exactly, reject before computing. Also, technically you probably _really_ need to reject only the datasets where the extreme values are not present. If some of the middle values do not...

I am happy to invest some energy into moving this forward, but I'd like to understand your use case better before delving into this. Are you interested only in producing...

Unfortunately, automatic datset generation for non-linear `brms` models is not currently supported (the error should be clearer on that). For most use cases, I'd strongly suggest writing your own generator...

I'll keep it open - we definitely should at least improve the error message. It is also possible that supporting non-linear models won't be that hard - we never really...