Dataset Viewer
Auto-converted to Parquet Duplicate
Search is not available for this dataset
env_id
int64
10
4.08k
env_params
dict
trajectories
array 3D
3,939
{ "advection_speed": 0.9611729491876803, "diffusion": 0.5527727564102366 }
[[[0.0,-0.1863124817609787,-0.3698880672454834,-0.5480560660362244,-0.7182762622833252,-0.8781997561(...TRUNCATED)
3,930
{ "advection_speed": 0.9658352169297278, "diffusion": 0.4220271341999307 }
[[[0.0,-0.05009763687849045,-0.0999264121055603,-0.14923124015331268,-0.19778411090373993,-0.2453963(...TRUNCATED)
138
{ "advection_speed": 0.04127364481693196, "diffusion": 0.17382035958375475 }
[[[1.6543612251060553e-24,0.03271523118019104,0.06536543369293213,0.09788806736469269,0.130225583910(...TRUNCATED)
3,329
{ "advection_speed": 0.8173721530805009, "diffusion": 0.03153744904417375 }
[[[0.0,-0.027070200070738792,-0.05470013618469238,-0.08343356102705002,-0.1137828379869461,-0.146214(...TRUNCATED)
1,742
{ "advection_speed": 0.43131835324797946, "diffusion": 0.23691488896515414 }
[[[0.0,-0.07542882114648819,-0.1494714319705963,-0.22079487144947052,-0.28817039728164673,-0.3505208(...TRUNCATED)
1,289
{ "advection_speed": 0.3228905051696195, "diffusion": 0.15037905368120738 }
[[[0.0,-0.05693181976675987,-0.11281028389930725,-0.1666054129600525,-0.21733349561691284,-0.2640784(...TRUNCATED)
3,542
{ "advection_speed": 0.8653385183403672, "diffusion": 0.36286966645935215 }
[[[0.0,-0.10979373008012772,-0.21897348761558533,-0.32691818475723267,-0.43299320340156555,-0.536545(...TRUNCATED)
1,606
{ "advection_speed": 0.4046715687101601, "diffusion": 0.11499260827718435 }
[[[0.0,0.011502516455948353,0.02316632866859436,0.03514817729592323,0.04759584739804268,0.0606443211(...TRUNCATED)
2,090
{ "advection_speed": 0.5179287293486837, "diffusion": 0.665535893321101 }
[[[0.0,0.011253022588789463,0.022394629195332527,0.03331086039543152,0.04388280585408211,0.053984455(...TRUNCATED)
484
{ "advection_speed": 0.12095395141140496, "diffusion": 0.5727659724623102 }
[[[0.0,-0.007799806073307991,-0.016049304977059364,-0.02518480457365513,-0.03561613708734512,-0.0477(...TRUNCATED)
End of preview. Expand in Data Studio

disco-ad

Trajectories used in the ICML 2026 paper Test-Time Generalization via Neural Operator Splitting (Serrano et al.).

Format

Each HDF5 file contains:

  • trajectories: shape (N, T, C, *spatial), float32
  • env_id: shape (N,), int64 — environment index for each trajectory
  • env_params/*: optional metadata mapping env_id back to PDE coefficients

Load via the project's train_generic.py / test_generic.py scripts:

from train.train_generic import GenericHDF5Dataset
ds = GenericHDF5Dataset(["train.h5"])

Or using huggingface_hub:

from huggingface_hub import hf_hub_download
local = hf_hub_download(repo_id="sogeeking/disco-ad", filename="train.h5", repo_type="dataset")

Code: https://github.com/LouisSerrano/disco-ball

Downloads last month
-