Obtaining predictions from multilevel model trace

[OriolAbril]

Links to the related topics:

Bug in fast sample posterior predictive?

I encountered a bug in pm.fast_sample_posterior_predictive(). Or maybe it is intended behavior that I don’t understand. Consider this simple model with two normals and three deterministics: import pymc3 as pm import numpy as np import xarray as xr import arviz as az with pm.Model() as model: n1 = pm.Normal('n1', 0, 1) d1 = pm.Deterministic('d1', n1) n2 = pm.Normal('n2', 0, 1) d2 = pm.Deterministic('d2', n2) d3 = pm.Deterministic('d3', d2) After sampling, all five var…

11.16 Rethinking Code

I’m going through the statistical rethinking notebooks and I’m not sure I quite understand what’s happening in cell Out[17] of chapter 11, 11.16 link with m11_4: pm.set_data({"actor_id": np.repeat(range(7), 4), "treat_id": list(range(4)) * 7}) p_post = pm.sample_posterior_predictive( trace_11_4, random_seed=RANDOM_SEED, var_names=["p"] )["p"] p_mu = p_post.mean(0).reshape((7, 4)) I’m not sure I understand what pm.set_data is doing in this particular case. I’m also not sure …

This last topic has inside it a link to a notebook which could be quite helpful alone