Contents Menu Expand Light mode Dark mode Auto light/dark, in light mode Auto light/dark, in dark mode Skip to content
RDR Eval
RDR Eval

Contents

  • Getting started
  • Example gallery
    • Relative density ratio for 1D Gaussians
    • Relative density ratio for 2D Gaussian mixtures
  • API reference
Back to top
View this page

RDR Eval¶

RDR Eval is a lightweight PyTorch package for estimating the relative density ratio

\[r(x) = \frac{2p(x)}{p(x)+q(x)}.\]

The package provides bounded neural ratio estimation, multiple divergence objectives, reproducible train/validation/test splitting, and validation-based early stopping.

Contents

  • Getting started
    • Installation
    • Basic usage
    • Validation and testing
  • Example gallery
  • API reference
    • Divergence
    • Trainer
    • Evaluation helpers

Indices and tables¶

  • Index

  • Module Index

  • Search Page

Next
Getting started
Copyright © 2026, Yuliang Xu, Yun Wei, and Li Ma
Made with Sphinx and @pradyunsg's Furo
On this page
  • RDR Eval
    • Indices and tables