Closed saucetray closed 1 year ago
I think it would help if you took a step back and wrote down your model, because it's not clear from this what your priors are and what you are trying to infer.
If you're unfamiliar with setting up bayesian models, which I suspect you might be, I would suggest working through Statistical Rethinking which is an excellent text book and also has free online lectures.
I am trying to use this library to do AB testing... I need to fit a prior before and I think using inference is fine. However, I can't retrieve the value for the scale of Gamma in
location
orscale
. I can optimize to find the location usinghistorical.optimize(location)
but that's it.I need to be able to do that and then use the values to run inference like this
However, this doesn't work because the sampler breaks since I am using scalars... So I thought the solution would be to then create more latent variables using the learned location and scale but I can only obtain their means for gamma and not their scales. How can I solve this?