CODES Benchmark#
CODES is an end-to-end benchmark for coupled ODE surrogates. It ships with curated datasets, reproducible tuning/training/evaluation scripts, and a comprehensive API so you can extend the stack with new models.
Note
A typical workflow follows Tune → Train → Evaluate. Use the links below to jump straight into the relevant guide, or follow the quickstart links in Getting Started to run a toy experiment on your machine.
Guided Quickstart — Getting Started walks you through cloning, installing, and running a smoke-test benchmark.
Benchmark Workflow — Running Benchmarks explains how tuning feeds training and how evaluations consolidate metrics.
Extend the Stack — Extending The Benchmark shows how to add datasets or surrogates without rewriting orchestration glue.
API Reference — API Reference Overview explains how the generated package docs are organized and links to each module.
Looking for a bird’s-eye view first? Start with the User Guide. Already configuring experiments or integrating your own model? Skip ahead to the API Reference. Either way, the sidebar mirrors the sections below so you are one click away from the next step.