In the arviz docs , there is this example using from_pymc3: trace = pm.sample(draws, chains=chains) prior = pm.sample_prior_predictive() posterior_predictive = pm.sample_posterior_predictive(trace) pm_data = az.from_pymc3( trace=trace, prior=prior, posterior_predictive=posterior_predict…

I’d recommend taking a look at A Hierarchical model for Rugby prediction — PyMC3 documentation and A Primer on Bayesian Methods for Multilevel Modeling — PyMC3 documentation notebooks

How to make InferenceData returned by sample() aware of the prior and posterior_predictive

Questions

DanWeitzenfeld September 8, 2021, 5:11pm 4

Thanks! It looks like extend and and the idata_orig argument to arviz.from_pymc3_predictions will do what I’m looking for.

1 Like

Topic		Replies	Views
Create InferenceData from trace and sampled posterior prediction Questions	2	1586	October 12, 2020
InferenceData incomplete Questions arviz	12	666	March 17, 2023
How to set prior in arviz InferenceData from pm.sample_prior_predictive()? Questions	2	988	April 22, 2021
Plotting Posterior Predictive using arviz Questions	2	662	August 12, 2021
Prior, prior predictive and sample_from_observed Development	0	868	April 18, 2020

How to make InferenceData returned by sample() aware of the prior and posterior_predictive

Related topics