Hi everybody! I’m working through Bayesian Modeling and Computation in Python book, and I ran into an issue trying to solve problem 1H23. I defined the model and ran it for five made-up observations (Y). I’d like to create a posterior predictive for 10 draws rather than 5. How should I go about it? …

In that case, you might just take the samples from your trace and use them directly. # how many credible data sets do we want? n_datasets = 4 # how big is each? n_obsperdataset = 10 # this is the number of 'flips' used in the model specification n = 1 # grab some samples from the trace thetas = …

How to change number of samples drawn by sample_posterior_predictive

Questions v3

Igor_K April 5, 2022, 2:08pm 5

I see - yes, that totally makes sense! Thank you!!

Topic		Replies	Views
Number of samples generated by `sample_posterior_predictive` Questions	3	1097	October 7, 2020
How to generate posterior predictive samples with size different than the observed variable? Questions	1	391	June 16, 2021
Example for out-of-sample prediction with posterior predictive sampling v5	8	3314	October 28, 2022
Posterior Predictive Checks Questions	1	614	May 19, 2019
Understanding shape of values returned by sample_posterior_predictive Questions	3	915	January 26, 2020

How to change number of samples drawn by sample_posterior_predictive

Related topics