Hi! I’m new to PyMC3, I wanted to say that it’s a great piece of software. I come from stan, and the transition has been pretty seamless! I had one question, about dimensions and posterior predictive checks. I’m using mini-batching ADVI to fit Peter Hoff’s AMEN model example in stan . The data I use…

I’ve got a solution! Based on this , I rewrote the code like so: Y = theano.shared(df.y.values) y = pm.Minibatch(df.y.values, 50) sender_full = theano.shared(df.sender.values - 1) sender = pm.Minibatch(df.sender.values - 1, 50) receiver_full = theano.shared(df.receiver.values - 1) receiver = pm.M…

Minibatch ADVI ppc dimensions

Questions

AlexAndorra August 10, 2020, 9:49am 5

Well done, and thanks for posting the solution

1 Like

Topic		Replies	Views
TypeError: object of type 'ADVI' has no len() when calling sample_posterior_predictive Questions	4	876	March 29, 2019
Minibatch when latent variable size depends on data dimension Questions	2	689	February 8, 2019
Minibatch variables not behaving as expected Questions	11	1568	March 14, 2018
Advi_minibatch is deprecated? Questions	4	1297	September 20, 2018
Getting Posterior Prediction of the data after using Minibatch v5	2	505	June 2, 2023

Minibatch ADVI ppc dimensions

Related topics