Thank you, Riccardo. I tried to correct the code like this:
with model:
pm.set_data({"control_data":z_test})
idata.extend(pm.sample_posterior_predictive(idata, var_names=["y_hat", "mu"]))
but I obtained the following error:
KeyError: ‘mu’
If I leave only y_hat, the code runs normally and I am able to extract the values, but if I add ‘mu’, it goes into error. ‘Mu’ is correctly present in the model above:
mu = pm.Deterministic("contributions", tt.stack(contributions).T, dims=['all_vars'])
How can I solve this issue?
Thank you!