coarse-grained-forecasting-benchmark/nonlinear-kg

2-D Nonlinear Klein-Gordon Conditional Forecasting

This dataset contains conditional ensembles of trajectories from the two-dimensional nonlinear Klein-Gordon equation. Each ensemble shares the same observed coarse initial state but differs in its hidden fine-scale initial conditions, producing multiple possible future trajectories.

The dataset accompanies an anonymous research submission.

Dataset

The observed state consists of 392 real-valued Fourier coefficients: 196 of the displacement (u) and 196 of the scaled velocity (u_t/4), recorded at 41 time points between (t=0) and (t=12).

The main file, data/trajectories_fine.npz, contains:

Array	Shape	Description
vecs_train	(80, 200, 41, 392)	Training ensembles
vecs_eval	(20, 200, 41, 392)	Validation and additional unused ensembles

The main experiments used 80 training initial conditions with 200 realizations each.

The file also contains group identifiers and time coordinates. All trajectory arrays are stored in float64.

Train, validation, and test conditions are specified in data/split.json and are disjoint. The dataset contains 80 training, 10 validation, and 10 test conditions.

Test reference ensembles

truth_bank/ contains 10,000 reference trajectories for each of the 10 test conditions. These are used to evaluate the predicted conditional distributions.

The test conditions are stored only in truth_bank/, not in vecs_eval.

Files and usage

• data/trajectories_fine.npz — coarse trajectories
• data/split.json — group-level data splits
• data/normalization.npz — training-derived normalization parameters
• truth_bank/group_<gid>.npz — test reference ensembles
• SHA256SUMS — file checksums

Download using:

hf download coarse-grained-forecasting-benchmark/nonlinear-kg --repo-type dataset --local-dir nonlinear-kg

For the accompanying evaluation code, set NLKG_DATA to nonlinear-kg/data and NLKG_TRUTH_BANK to nonlinear-kg/truth_bank.

License

Creative Commons Attribution 4.0 International (CC BY 4.0).