Welcome to pyblip’s documentation!

pyblip is a python implementation of BLiP (Bayesian Linear Programming), a method to jointly localize and detect signals. Given nearly any Bayesian model, pyblip is designed so that BLiP can wrap on top of the Bayesian model in only one or two lines of code to enhance its power. Check out the Tutorials section for more examples!

Contents:

• Getting Started
  • Installation
    • Performance improvement with CBC
    • Installation issues
  • Minimal example
• Tutorials
  • 1. Introduction
  • 2. Variable selection
    • 2.1 Generating synthetic data
    • 2.2 Linear Spike-and-Slab model + BLiP
    • 2.3 SuSiE + BLiP
    • 2.4 Probit spike-and-slab model + BLiP
    • 2.5 Purity thresholding
    • 2.6 Hierarchical priors to ensure error control
  • 3. Changepoint detection
  • 4. Point-source detection
  • 5. Inputs for arbitrary problems
    • 5.1 Posterior samples for discrete problems
    • 5.2 Posterior samples for continuous problems
    • 5.3 Advanced usage: directly specifying candidate groups
• API Reference
  • BLiP
  • Creating Candidate Groups
    • Variable selection / discrete locations
    • Continuous sets of locations
    • Edge clique cover
    • Weight functions
  • Bayesian Samplers
    • Linear Spike and Slab
    • Probit Spike and Slab
    • Neuronized Priors
    • Changepoint Detection
• pyblip source
• blipr source
• BLiP paper (arXiv)

Indices and tables

• Index

• Module Index

• Search Page