Dataset Viewer
Auto-converted to Parquet Duplicate
sim_id
stringclasses
242 values
time_id
int64
0
3.95k
2062.h5
1,740
4312.h5
370
10125.h5
1,970
2437.h5
3,520
7500.h5
3,950
4031.h5
2,600
8156.h5
1,330
6656.h5
2,700
8625.h5
1,600
9937.h5
2,770
6187.h5
1,160
6656.h5
3,160
2718.h5
660
11343.h5
2,720
6000.h5
1,120
2062.h5
3,020
8812.h5
740
2062.h5
1,930
8156.h5
2,750
4500.h5
2,080
9281.h5
2,970
5062.h5
1,040
3093.h5
3,750
10406.h5
290
4781.h5
1,700
10875.h5
2,180
6093.h5
1,560
5062.h5
1,830
10125.h5
350
4312.h5
990
8718.h5
1,910
9937.h5
2,950
4312.h5
2,950
11156.h5
1,870
8531.h5
1,220
3656.h5
1,790
4687.h5
2,060
8718.h5
2,520
8531.h5
1,270
8718.h5
3,810
11812.h5
1,790
10875.h5
1,000
5062.h5
2,930
8625.h5
200
3656.h5
3,500
9468.h5
3,490
11718.h5
250
2062.h5
810
9000.h5
180
2156.h5
520
2062.h5
490
8156.h5
3,470
6000.h5
3,520
9281.h5
2,310
3843.h5
2,430
3000.h5
810
11531.h5
1,560
3093.h5
290
10312.h5
3,000
10218.h5
370
4031.h5
1,700
6656.h5
430
10218.h5
520
10875.h5
3,560
2062.h5
2,930
5156.h5
3,740
6187.h5
810
11250.h5
990
3843.h5
2,350
8343.h5
2,470
1875.h5
430
9468.h5
2,620
6187.h5
2,200
6656.h5
350
11062.h5
3,270
10781.h5
3,410
5906.h5
250
10687.h5
1,410
7031.h5
700
2812.h5
540
2718.h5
2,000
11437.h5
2,370
11437.h5
1,750
10687.h5
2,810
4875.h5
3,560
8906.h5
3,390
6093.h5
3,540
10218.h5
220
9468.h5
790
2718.h5
2,740
9656.h5
3,700
9187.h5
3,290
10500.h5
2,970
4593.h5
1,830
8156.h5
3,310
2718.h5
680
7031.h5
990
3843.h5
3,430
10593.h5
2,850
9000.h5
1,600
End of preview. Expand in Data Studio

RealPDEBench logo

RealPDEBench

HF Dataset arXiv Website & Docs Codebase License: CC BY-NC 4.0

RealPDEBench is a benchmark of paired real-world measurements and matched numerical simulations for complex physical systems. It is designed for spatiotemporal forecasting and sim-to-real transfer evaluation on real data.

This Hub repository (AI4Science-WestlakeU/RealPDEBench) is the release repo for RealPDEBench.

RealPDEBench overview figure

Figure 1. RealPDEBench provides paired real-world measurements and matched numerical simulations for sim-to-real evaluation.

What makes RealPDEBench different?

  • Paired real + simulated data: each scenario provides experimental measurements and corresponding CFD/LES simulations.
  • Real-world evaluation: models are evaluated on real trajectories to quantify the sim-to-real gap.
  • Multi-modal mismatch: simulations include additional unmeasured modalities (e.g., pressure, species fields), enabling modality-masking and transfer strategies.

Data sources (high level)

  • Fluid systems (cylinder, controlled_cylinder, fsi, foil):
    • Real: Particle Image Velocimetry (PIV) in a circulating water tunnel
    • Sim: CFD (2D finite-volume + immersed-boundary; 3D GPU solvers depending on scenario)
  • Combustion (combustion):
    • Real: OH* chemiluminescence imaging (high-speed)
    • Sim: Large Eddy Simulation (LES) with detailed chemistry (NH3/CH4/air co-firing)

Scenarios (5)

Scenario Real data (measured) Numerical data (simulated) Frames / trajectory Spatial grid (after sub-sampling) HDF5 trajectories (real / numerical)
cylinder velocity (u,v) (u,v,p) 3990 64×128 92 / 92
controlled_cylinder (u,v) (u,v,p) (+ control params in filenames) 3990 64×128 96 / 96
fsi (u,v) (u,v,p) 2173 64×64 51 / 51
foil (u,v) (u,v,p) 3990 64×128 98 / 99
combustion OH* chemiluminescence intensity (1 channel) intensity surrogate (1) + 15 simulated fields 2001 128×128 30 / 30

Total trajectories (HDF5 files): ~735 (≈367 real + ≈368 numerical).

Physical parameter ranges (real experiments)

Scenario Key parameters (real)
cylinder Reynolds number (Re): 1800–12000
controlled_cylinder (Re): 1781–9843; control frequency (f): 0.5–1.4 Hz
fsi (Re): 3272–9068; mass ratio (m^*): 18.2–20.8
foil angle of attack (\alpha): 0°–20°; (Re): 2968–17031
combustion CH4 ratio: 20–100%; equivalence ratio (\phi): 0.75–1.3

Data format on the Hub

RealPDEBench stores complete trajectories in HuggingFace Arrow format, with separate JSON index files for train/val/test splits. This enables dynamic N_autoregressive support at runtime.

Each scenario contains:

  • Trajectory data: hf_dataset/{real,numerical}/ — Arrow files with complete time series
  • Index files: hf_dataset/{split}_index_{type}.json — maps sample indices to (sim_id, time_id)
  • test_mode metadata: {in_dist,out_dist,remain}_params_{type}.json

Repository layout:

{repo_root}/
  cylinder/
    in_dist_test_params_real.json
    out_dist_test_params_real.json
    remain_params_real.json
    in_dist_test_params_numerical.json
    out_dist_test_params_numerical.json
    remain_params_numerical.json
    hf_dataset/
      real/                           # Arrow: complete trajectories (92 files)
        data-*.arrow
        dataset_info.json
        state.json
      numerical/                      # Arrow: complete trajectories
        data-*.arrow
        dataset_info.json
        state.json
      train_index_real.json           # Index: [{"sim_id": "xxx.h5", "time_id": 0}, ...]
      val_index_real.json
      test_index_real.json
      train_index_numerical.json
      val_index_numerical.json
      test_index_numerical.json
  fsi/
    ...  (same structure)
  controlled_cylinder/
    ...  (same structure)
  foil/
    ...  (same structure)
  combustion/
    ...  (same structure)

How to download only what you need

For large data, use snapshot_download(..., allow_patterns=...) to avoid pulling the full repository.

import os
from huggingface_hub import snapshot_download
from datasets import load_from_disk

repo_id = "AI4Science-WestlakeU/RealPDEBench"
os.environ["HF_HUB_DISABLE_XET"] = "1"
local_dir = snapshot_download(
    repo_id=repo_id,
    repo_type="dataset",
    allow_patterns=["fsi/**"],  # example: download only the FSI folder
    endpoint="https://hf-mirror.com",
)

# Load trajectory data
trajectories = load_from_disk(os.path.join(local_dir, "fsi", "hf_dataset", "real"))
print(f"Loaded {len(trajectories)} trajectories")
print(trajectories[0].keys())  # sim_id, u, v, shape_t, shape_h, shape_w

Using the RealPDEBench loaders (recommended)

For automatic train/val/test splitting and dynamic N_autoregressive support, use the provided dataset loaders:

from realpdebench.data.fluid_hf_dataset import FSIHFDataset

dataset = FSIHFDataset(
    dataset_name="fsi",
    dataset_root="/path/to/data",
    dataset_type="real",
    mode="test",
    N_autoregressive=10,  # Dynamic! Works with any value
)

input_tensor, output_tensor = dataset[0]
print(f"Input shape: {input_tensor.shape}")   # (20, H, W, 2)
print(f"Output shape: {output_tensor.shape}") # (200, H, W, 2) = 20 × 10

Schema (columns)

Fluid datasets (cylinder, controlled_cylinder, fsi, foil)

  • Keys (each row = one complete trajectory):
    • sim_id (string): trajectory file name (e.g., 10031.h5)
    • u, v (bytes): float32 arrays of shape (T_full, H, W)complete time series
    • p (bytes): float32 array (T_full, H, W) (numerical splits only)
    • shape_t (int): complete trajectory length (e.g., 3990, 2173)
    • shape_h, shape_w (int): spatial dimensions

Combustion dataset (combustion)

  • Keys (each row = one complete trajectory):
    • sim_id (string): e.g., 40NH3_1.1.h5
    • observed (bytes): float32 array (T_full, H, W)complete time series
    • numerical (bytes): float32 array (T_full, H, W, 15) (numerical splits only)
    • numerical_channels (int): number of numerical channels (15)
    • shape_t (int): complete trajectory length (e.g., 2001)
    • shape_h, shape_w (int): spatial dimensions

Index files (JSON)

Each split has an index file mapping sample indices to trajectory positions:

[
  {"sim_id": "10031.h5", "time_id": 0},
  {"sim_id": "10031.h5", "time_id": 20},
  {"sim_id": "10031.h5", "time_id": 40},
  ...
]

Data size

  • Total: ~210GB across all scenarios
  • Largest shard file: ~0.5GB (well below the Hub's recommended <50GB per file)
  • Total file count: ~550 files (well below the Hub's recommended <100k files per repo)

Per-scenario totals:

Scenario real numerical Total
cylinder 23GB 34GB 57GB
controlled_cylinder 24GB 36GB 59GB
fsi 6GB 11GB 17GB
foil 24GB 37GB 61GB
combustion 1GB 15GB 16GB
Total 78GB 133GB ~210GB

Recommended benchmark protocols

RealPDEBench supports three standard training paradigms (all evaluated on real-world data):

  • Simulated training (numerical only)
  • Real-world training (real only)
  • Simulated pretraining + real finetuning

License

This dataset is released under CC BY‑NC 4.0 (non‑commercial). Please credit the authors and the benchmark paper when using the dataset.

Citation

If you find our work and/or our code useful, please cite us via:

@misc{hu2026realpdebenchbenchmarkcomplexphysical,
      title={RealPDEBench: A Benchmark for Complex Physical Systems with Real-World Data}, 
      author={Peiyan Hu and Haodong Feng and Hongyuan Liu and Tongtong Yan and Wenhao Deng and Tianrun Gao and Rong Zheng and Haoren Zheng and Chenglei Yu and Chuanrui Wang and Kaiwen Li and Zhi-Ming Ma and Dezhi Zhou and Xingcai Lu and Dixia Fan and Tailin Wu},
      year={2026},
      eprint={2601.01829},
      archivePrefix={arXiv},
      primaryClass={cs.LG},
      url={https://arxiv.org/abs/2601.01829}, 
}

Contact

AI for Scientific Simulation and Discovery Lab, Westlake University
Maintainer: westlake-ai4s (Hugging Face)
Org: AI4Science-WestlakeU

Downloads last month
676

Paper for AI4Science-WestlakeU/RealPDEBench