Configuration

This page serves as an interactive config generator for the CODES Benchmark. Fill in the fields below to create a custom config.yaml file.


Config Generator

Provide a unique identifier for this training run (e.g., "my_run123").

Dataset

Specify which dataset to use, how to transform it, and additional parameters.

Surrogates to Include

Select surrogates and set their parameters.

Miscellaneous Settings

Configure devices, verbosity, and random seed for reproducibility.

Models to Train (Benchmark Tasks)

Enable and configure interpolation, extrapolation, sparse data usage, uncertainty quantification, and batch size scaling.

Interpolation

Extrapolation

Sparse Training Data

Uncertainty Quantification

Batch Size Scaling

Evaluations During Benchmark

Select which evaluations to perform and record.