mcreweight documentation

This section documents the available reweighting methods and how their training objectives differ.

Contents

• Reweighting methods
  • Overview
  • Method-by-method behavior
  • Data visualization and diagnostics
  • Loss function and update mechanics
  • Validation and early stopping
  • Optuna hyperparameter optimization
  • Feature transformations
  • Main differences with hep_ml
  • Which method to use
• Configuration reference
  • run-reweight
  • apply-weights
  • Environment variables
  • Legacy key aliases
• API reference
  • Pipelines
  • Training helpers
  • Main reweighter classes
  • Folding classes
  • Plotting utilities
• Examples
  • Fixture files
  • Example 1: train reweighters
  • Example 2: apply a trained model
  • Example 3: throughput and memory sweep
  • Reading the outputs