CODES Configuration

Training ID

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.

Dataset name (e.g. 'osu2008')

Apply log10 transform?

Tolerance (e.g. 1e-30)

Normalization

Use optimal model hyperparameters

Subset Factor (e.g. 1)

Surrogates to Include

Select surrogates and set their parameters.

Fully Connected Neural Network

Latent Neural ODE

Add model

Deep Operator Network

Latent Polynomial Network

Miscellaneous Settings

Configure devices, verbosity, and random seed for reproducibility.

Devices

Random Seed

Verbose output (does not change much currently)

Models to Train (Benchmark Tasks)

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

Interpolation

Enable Interpolation

Use every n-th timestep:

Extrapolation

Enable Extrapolation

Cutoff at timesteps...

Sparse Training Data

Enable Sparse Training Data

Use every n-th sample:

Uncertainty Quantification

Enable Uncertainty Quantification

Ensemble size:

Batch Size Scaling

Enable Batch Size Scaling

Evaluations During Benchmark

Select which evaluations to perform and record.

Save and plot losses

Save and plot dynamic accuracy

Save and plot timing

Measure Compute performance

Compare the surrogates with each other in the benchmark?

Configuration

Config Generator