/potentials

Potential functions for sampling or optimization

Primary LanguageJupyter NotebookMIT LicenseMIT

Benchmark potential functions in PyTorch

This package provides potential functions to benchmark sampling and optimization algorithms.

Synthetic potentials include:

  • Standard Gaussian
  • Diagonal Gaussian (several variants)
  • Block diagonal Gaussian (several variants)
  • Full rank Gaussian (several variants)
  • Multimodal (several variants)
  • Funnel
  • Rosenbrock

Real-world potentials include:

  • Biochemical oxygen demand model
  • German credit model
  • Predator prey system model