PyBEAM: A Bayesian approach to parameter inference for a wide class of binary evidence accumulation models

PyBEAM (Bayesian Evidence Accumulation Models) is a Python package designed to rapidly fit two-threshold, binary choice models to choice-RT data using Bayesian inference methods. For a full description of its design, see the publication (https://psyarxiv.com/ax36b/). For access to the package code and other files, see the PyBEAM github (https://github.com/murrowma/pybeam/). To learn how to use PyBEAM, see the Precoded tutorials and Custom tutorials tabs for step-by-step instructions.

Note

This project is under active development.

Contents

• Installation
  • System requirements and package dependencies
  • Installing PyBEAM
  • Learning to use PyBEAM
  • Custom models
• Precoded functions
  • pybeam.precoded.simulate()
  • pybeam.precoded.likelihood()
  • pybeam.precoded.loglikelihood()
  • pybeam.precoded.plot_rt()
  • pybeam.precoded.inference()
  • pybeam.precoded.plot_idata()
  • pybeam.precoded.summary()
• Precoded tutorials
  • Tutorial 1 - Models available in PyBEAM
  • Tutorial 1a - simpleDDM
  • Tutorial 1b - fullDDM
  • Tutorial 1c - leakage
  • Tutorial 1d - changing_thresholds
  • Tutorial 1e - UGM
  • Tutorial 1f - UGM_flip
  • Tutorial 2 - Using your model
  • Tutorial 3 - Parameter inference
  • Tutorial 4 - Parameter inference for multiple condition models
  • Tutorial 5 - Fixing model parameters
  • Tutorial 6 - Fitting models using optional inputs
  • Tutorial 7 - Fitting experimental data
• Custom functions
  • pybeam.custom.functions_test()
  • pybeam.custom.simulate()
  • pybeam.custom.likelihood()
  • pybeam.custom.loglikelihood()
  • pybeam.custom.plot_rt()
  • pybeam.custom.inference()
  • pybeam.custom.plot_idata()
  • pybeam.custom.summary()
• Custom tutorials
  • Custom Tutorial 1: Creating a custom model version of the simple DDM model
  • Custom Tutorial 1a: Creating a custom model version of the UGM_flip model
  • Custom Tutorial 2: Using your model
  • Custom Tutorial 3: Parameter inference