Welcome to NumPyro Tutorials!

Tutorials:

• Bayesian Regression Using NumPyro
  • Tutorial Outline:
  • Dataset
  • Regression Model to Predict Divorce Rate
  • Regression Model with Measurement Error
  • Recording the number of gradient evaluations used by NUTS
  • References
• Time Series Forecasting
  • Data
  • Model
  • Inference
  • Forecasting
  • Acknowledgements
  • References
• Solving differential equations (ODEs) for multiple initial conditions.
  • Model
  • Dataset
  • Perform MCMC.
  • Run predictions.
• Bayesian Imputation
  • Dataset
  • Prepare data
  • Modelling
  • Sampling
  • Prediction
  • References
• Ordinal Regression
  • Data Generation
  • Improper Prior
  • Proper Prior
  • Principled prior with Dirichlet Distribution
• Bayesian Hierarchical Linear Regression
  • 1. Understanding the task
  • 2. Modelling: Bayesian Hierarchical Linear Regression with Partial Pooling
  • 3. Fitting the model
  • 4. Checking the model
  • 5. Add layer to model hierarchy: Smoking Status
  • Wrap-up
  • References
• Bayesian Imputation for Missing Values in Discrete Covariates
  • MAR conditional on outcome
  • MNAR conditional on covariate
• Automatic rendering of NumPyro models
  • A Simple Example
  • Tweaking the visualization
  • Rendering the parameters
  • Distribution and Constraint annotations
  • Rendering deterministic sites
  • Rendering neural network’s parameters
  • Overlapping non-nested plates
• Composing Gaussian Processes with NumPyro using GPJax
  • Data simulation
  • Building the GP inside the NumPyro model
  • MCMC inference
  • Posterior predictions
  • Visualisation
• Truncated and folded distributions
  • Table of contents
  • Setup
  • 1. What are truncated distributions?
  • 2. What is a folded distribution?
  • 3. Sampling from truncated and folded distributions
  • 4. Ready to use truncated and folded distributions
  • 5. Building your own truncated distribution
  • References and related material

Indices

• Index