Nikhil Sarin
Nikhil Sarin
Better than what? Sure, send through. I do have some of my own likelihoods though. You can always just use your own likelihood in fitting.
There's plenty of other ways it's been/can be done. I like to use a standard gaussian for detections and add the probability that the true value is below the upper...
Been in the inference game for a while :P
@GPLamb #277 includes an overhaul to extinction models to do av/rv for host and mw separately (example in the docs) and a non-detection likelihood example. A bunch of other stuff...
closing this, @GPLamb you can reopen if necessary