Search is not available for this dataset
pos sequencelengths 350 350 | mat sequencelengths 496 1.4k | n_particles int64 496 1.4k | n_steps int64 350 350 | friction_angle sequencelengths 1 1 |
|---|---|---|---|---|
[[[0.5451056957244873,0.41256996989250183],[0.720942497253418,0.4259949028491974],[0.525687932968139(...TRUNCATED) | [6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6(...TRUNCATED) | 832 | 350 | [
34.10598240865633
] |
[[[0.5139346718788147,0.5805840492248535],[0.582686722278595,0.6127313375473022],[0.5479996800422668(...TRUNCATED) | [6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6(...TRUNCATED) | 730 | 350 | [
28.92145985835677
] |
[[[0.38689252734184265,0.678761899471283],[0.5085696578025818,0.6903204321861267],[0.373455703258514(...TRUNCATED) | [6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6(...TRUNCATED) | 496 | 350 | [
32.31041216458434
] |
[[[0.4638594388961792,0.6778674125671387],[0.694644033908844,0.6950327157974243],[0.4383738040924072(...TRUNCATED) | [6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6(...TRUNCATED) | 1,396 | 350 | [
45.37040831812692
] |
[[[0.5978336930274963,0.49911487102508545],[0.5464008450508118,0.47499293088912964],[0.5911015868186(...TRUNCATED) | [6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6(...TRUNCATED) | 1,369 | 350 | [
28.16303931632747
] |
MPM 2D Sand Dataset
This dataset includes trajectories of a sand cube simulated using Taichi's MPM framework. The initial position, velocity, size of the cube are randomized.
Installation
To download the dataset you don't need an account for huggingface. You only need to install the package by running
pip install datasets
Loading Dataset
from datasets import load_dataset
ds = load_dataset("chhsiao93/mpm_sand_2d_single_cube", split='train')
n = 0 # the first trajectory
trajectory = ds[n]
for key, _ in trajectory.items():
print(key) # show all the keys in the dictionary
You can also load the data as an iterable table. Here is an example:
# loading as an iterable dataset
ds = load_dataset("chhsiao93/mpm_sand_2d_single_cube", split='train', streaming=True)
# iterate through each trajectory and save it as an npz file
for i, traj in enumerate(iter(ds)):
temp = {}
for key, value in traj.items():
temp[key] = np.array(value)
np.savez(f'trajectory{i}.npz', **temp)
print(f'saving trajectory{i} as npz')
- Downloads last month
- 6