| | --- |
| | license: cc-by-nc-4.0 |
| | --- |
| | |
| | # Short Description |
| |
|
| | Based on the incompressible Navier-Stokes equations, this dataset contains trajectories starting from double shear layer initial conditions, see https://arxiv.org/abs/2405.19101. |
| |
|
| | # Dimensions |
| |
|
| | The assembled NetCDF file has a **single** variable called *velocity* with dimensionality |
| | - 40000 (number of trajectories) |
| | - 21 (time steps) |
| | - 2 (horizontal velocity, vertical velocity) |
| | - 128 (x-dim) |
| | - 128 (y-dim) |
| |
|
| | It was simulated on the unit square up to T=1 and saved as uniformly spaced in space and time. |
| |
|
| | # Train/Val/Test-split |
| |
|
| | 39640/120/240 trajectories |
| |
|
| | # Download & Assembly |
| |
|
| | The dataset can be downloaded, e.g., via `huggingface-cli download`. |
| |
|
| | After download, the chunked data can be assembled into a single NetCDF file using the provided `assemble_data.py` script. |
| | Use it as follows: |
| | ```bash |
| | python assemble_data.py --input_dir . --output_file NS-SL.nc |
| | ``` |
