DeepLag datasets and pretrained checkpoints
Authors/Creators
Description
Datasets and pretrained checkpoints accompanying DeepLag: Discovering Deep Lagrangian Dynamics for Intuitive Fluid Prediction (https://arxiv.org/abs/2402.02425; code: https://github.com/thuml/DeepLag). Includes bounded Navier-Stokes simulation, ocean-current data, 3D smoke data, and the complete supplied model checkpoints.
Extract 3D_Smoke.tar and checkpoints.tar at the repository root. Place re0.25_4c_gray_10000.npy and boundary_4c_rot.npy in data/Bounded_NS/. Place sea_kuroshio.npy, sea_kuroshio_mean.npy, and land_mask_reconstructed.npy in data/Ocean_Current/. The mask is float32 with shape (180, 300), land=1 and sea=0. Ocean arrays are processed aggregates with shape (3650, 180, 300, 5); use an explicit mask loader instead of the original raw-frame mask-reading routine. The complete bounded-flow array is provided without duplicate split files.
Licensing: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/) applies to the team-generated datasets and model weights to the extent that the team holds the relevant rights. Third-party rights and applicable upstream terms are not superseded. Ocean-current data retain the Copernicus Marine Service terms: https://marine.copernicus.eu/user-corner/service-commitments-and-licence. Generated using E.U. Copernicus Marine Service Information; source: Global Ocean Physics Reanalysis, https://doi.org/10.48670/moi-00021.
Files
SHA256SUMS.txt
Files
(41.0 GB)
| Name | Size | |
|---|---|---|
|
md5:4e7fa42a9b1a5e24d7bc10d2eb1ed9f6
|
12.6 GB | Download |
|
md5:7313c0e77487616eb8441b024428148e
|
262.3 kB | Download |
|
md5:326cea7150ff9b84048218afe3013691
|
3.5 GB | Download |
|
md5:7eef4f403abd8baffbb15d33b178ad03
|
216.1 kB | Download |
|
md5:b2d4062ec0401f75aa3c55a576ca9800
|
21.0 GB | Download |
|
md5:578c46945fc61a23bc97f8eb4498b255
|
3.9 GB | Download |
|
md5:ebda4ca1f10c817bdebec90305dfb32a
|
2.2 MB | Download |
|
md5:e49bfdf06013d383ee250094d9460a8d
|
509 Bytes | Preview Download |
|
md5:526e6b6687aa094f5a33f70dbc7b31f4
|
370.5 kB | Preview Download |