All that likelihood with PyMC3
summary
Shows how to extract actual likelihoods from a PyMC3 model: extract component logp and evaluate model_logp sum at test points, visualize likelihood surfaces and log‑likelihood slices, and compare how NUTS versus variational inference use that quantity for inference and model comparison. Demonstrates practical pitfalls — reparameterization changes volume, priors wash out with more data, and label-switching in mixtures requires transformations — using PyMC3 (Theano back end) and provided notebooks that reproduce the examples.
AI-written summary · about
chapters (6)
- 0:00PyMC3 introduction
- 10:46Probability and random variables
- 20:30Probability to likelihood
- 30:52Likelihoods and priors
- 37:25Model parameterization
- 39:14PyMC3 and Pyro Q&A
related talks
44mAn introduction to PyMC3
39mProbablistic Programming Data Science with PyMC3
57mA Quickstart Guide to PyMC3
01h 27mProbabilistic Python: An Introduction to Bayesian Modeling with PyMC
01h 31mProbabilistic Programming and Bayesian Inference in Python
43mProbabilistic Programming in the Real World
01h 26mProbabilistic Programming and Bayesian Computing with PyMC
42mBayesian Non-parametric Models for Data Science using PyMC3