Probabilistic Programming and Bayesian Inference in Python
summary
Hands-on introduction to Bayesian inference using PyMC3 and Theano, showing concrete code: build priors, feed observed data, run MCMC traces (defaults 2 chains, examples use 500–2,500 samples), and inspect trace plots and posterior predictive checks. Demonstrates that a loose prior lets the sampler recover true parameters from synthetic data while a restrictive prior biases estimates, and sketches using hierarchical models (mixed effects) for grouped data and when pooling fails.
AI-written summary · about
chapters (5)
- 0:00Start
- 10:22Bayesian Inference vs Frequentist
- 20:25Probabilistic Programming
- 56:10Hierarchical Linear Regression
- 58:34Mixed Effects Modeling
related talks
39mProbablistic Programming Data Science with PyMC3
44mAn introduction to PyMC3
01h 27mProbabilistic Python: An Introduction to Bayesian Modeling with PyMC
57mA Quickstart Guide to PyMC3
42mBayesian Non-parametric Models for Data Science using PyMC3
43mProbabilistic Programming in the Real World
01h 00mA Hitchhiker's Guide to designing a Bayesian library in Python
01h 26mProbabilistic Programming and Bayesian Computing with PyMC