Hi, I am trying to switch my model from pm.sample() to pm.fit() for performance reasons. I have gotten my model to sample well with pm.sample() on a subset of my data–meaning there are 0 divergences, the r_hats are all under 1.01, and the posteriors are reasonable–but my full dataset is much too l…

Transitioning from pm.sample() to pm.fit()

jessegrabowski May 28, 2025, 1:28pm 15

Can you post you model, or something similar to it? I’m curious to see if there’s any exotic features that would cause ADVI to just totally fail

Topic		Replies	Views
NaN occurred in optimization at first Iteration with ADVI Questions	3	764	February 18, 2020
NaN occurred in optimization with ADVI Questions	5	4619	August 6, 2019
ADVI changes start values to NaN Questions	2	895	January 18, 2021
How to use pymc3.fit() method Questions	16	3836	May 7, 2018
NaN occurred in optimization with NUTS Questions	8	4719	May 29, 2018