My model, with less than 100 parameters and with ~5000 observable samples nicely… But crashes just after samping ends… [image] I guess that this occurs when computing end of sampling tests (to detect diverging samples for instance) where pymc3 uses a lot of memory… Can we disable these tests so…

Which PyMC3 and ArviZ version are you using? idata_kwargs was introduced on version 3.9, so I’d strongly recommend updating to the latest release. [image] hwassner: But the memory is still full and the problem arise later (outside pymc). So it doesn’t really solve my problem. Yes, the conver…

How to disable convergence checks?

Questions

OriolAbril September 16, 2020, 4:23pm 2

I believe the answers to this question will solve the issue. To solve the memory error, disabling the convergence checks should not be necessary, only avoiding the storage of pointwise log likelihood data.

Topic		Replies	Views
How can I terminate pm.sample when a convergence criterion is met (e.g. r_hat < 1.05) v5	3	563	March 14, 2023
"log_likelihood" not found in InferenceData Questions	5	3833	June 21, 2023
Bayesian regression: crashes and performance issues with large datasets and many parameters with different priors? Questions	7	1438	January 27, 2021
Memory spike at the end of the MCMC sampling Questions	10	2486	July 15, 2022
Compute_log_likelihood RAM usage is huge version agnostic arviz	4	427	September 27, 2023

How to disable convergence checks?

Related topics