metadata

license: apache-2.0
language:
  - en
tags:
  - datasets
  - machine-learning
  - deep-learning
  - physics-modeling
  - scientific-ML
  - material-point-method
  - MPM
  - smooth-particle-hydrodynamics
  - SPH
  - Lagrangian-dynamics
pretty_name: MPM-Verse-Large
size_categories:
  - 100K<n<1M

MPM-Verse-MaterialSim-Large

Dataset Summary

This dataset contains Material-Point-Method (MPM) simulations for various materials, including water, sand, plasticine, and jelly. Each material is represented as point-clouds that evolve over time. The dataset is designed for learning and predicting MPM-based physical simulations. The dataset is rendered using five geometric models - Stanford-bunny, Spot, Dragon, Armadillo, and Blub. Each setting has 10 trajectories per object.

Supported Tasks and Leaderboards

The dataset supports tasks such as:

Physics-informed learning
Point-cloud sequence prediction
Fluid and granular material modeling
Neural simulation acceleration

Dataset Structure

Materials and Metadata

Due to the longer duration, water and sand are split into multiple files for rollout_full and train. rollout_full represents the rollout trajectory over the full-order point-cloud, while rollout is on a sample size of 2600. The first 40 trajectories are used in the train set, and the remaining 10 are used in the test set.

Dataset Characteristics

Material	# of Trajectories	Duration	Time Step (dt)	Shapes	Train Sample Size
Water3DNCLAW	50	1000	5e-3	Blub, Spot, Bunny, Armadillo, Dragon	2600
Sand3DNCLAW	50	500	2.5e-3	Blub, Spot, Bunny, Armadillo, Dragon	2600
Plasticine3DNCLAW	50	200	2.5e-3	Blub, Spot, Bunny, Armadillo, Dragon	2600
Jelly3DNCLAW	50	334	7.5e-3	Blub, Spot, Bunny, Armadillo, Dragon	2600
Contact3DNCLAW	50	600	2.5e-3	Blub, Spot, Bunny	2600

Dataset Files

Each dataset file is a dictionary with the following keys:

`train.obj/test.pt`

particle_type (list): Indicator for material (only relevant for multimaterial simulations). Each element has shape [N] corresponding to the number of particles in the point-cloud.
position (list): Snippet of past states, each element has shape [N, W, D] where:
- N: Sample size
- W: Time window (6)
- D: Dimension (2D or 3D)
n_particles_per_example (list): Integer [1,] indicating the size of the sample N
output (list): Ground truth for predicted state [N, D]

`rollout.pt/rollout_full.pt`

position (list): Contains a list of all trajectories, where each element corresponds to a complete trajectory with shape [N, T, D] where:
- N: Number of particles
- T: Rollout duration
- D: Dimension (2D or 3D)

Metadata Files

Each dataset folder contains a metadata.json file with the following information:

bounds (list): Boundary conditions.
default_connectivity_radius (float): Radius used within the graph neural network.
vel_mean (list): Mean velocity of the entire dataset [x, y, (z)] for noise profiling.
vel_std (list): Standard deviation of velocity [x, y, (z)] for noise profiling.
acc_mean (list): Mean acceleration [x, y, (z)] for noise profiling.
acc_std (list): Standard deviation of acceleration [x, y, (z)] for noise profiling.

Downloading the Dataset

from huggingface_hub import hf_hub_download, snapshot_download

files = ['train.obj', 'test.pt', 'rollout.pt', 'metadata.json', 'rollout_full.pt']

train_dir = hf_hub_download(repo_id=params.dataset_rootdir, repo_type='dataset', filename=os.path.join('Jelly3DNCLAW', files[0]), cache_dir="./dataset_mpmverse")
test_dir = hf_hub_download(repo_id=params.dataset_rootdir, repo_type='dataset', filename=os.path.join('Jelly3DNCLAW', files[1]), cache_dir="./dataset_mpmverse")
rollout_dir = hf_hub_download(repo_id=params.dataset_rootdir, repo_type='dataset', filename=os.path.join('Jelly3DNCLAW', files[2]), cache_dir="./dataset_mpmverse")
metadata_dir = hf_hub_download(repo_id=params.dataset_rootdir, repo_type='dataset', filename=os.path.join('Jelly3DNCLAW', files[3]), cache_dir="./dataset_mpmverse")
rollout_full_dir = hf_hub_download(repo_id=params.dataset_rootdir, repo_type='dataset', filename=os.path.join('Jelly3DNCLAW', files[4]), cache_dir="./dataset_mpmverse")

Processing Train

import torch
import pickle

with open("path/to/train.obj", "rb") as f:
  data = pickle.load(f)

positions = data["position"][0]
print(positions.shape)  # Example output: (N, W, D)

Processing Rollout

import torch
import pickle

with open("path/to/rollout_full.obj", "rb") as f:
  data = pickle.load(f)

positions = data["position"]
print(len(positions))  # Example output: 50
print(positions.shape) # Example output: (N, T, 3)

Example Simulations

Citation

If you use this dataset, please cite:

@article{viswanath2024reduced,
  title={Reduced-Order Neural Operators: Learning Lagrangian Dynamics on Highly Sparse Graphs},
  author={Viswanath, Hrishikesh and Chang, Yue and Berner, Julius and Chen, Peter Yichen and Bera, Aniket},
  journal={arXiv preprint arXiv:2407.03925},
  year={2024}
}

Source

The 3D datasets (e.g., Water3D, Sand3D, Plasticine3D, Jelly3D, RigidCollision3D, Melting3D) were generated using the NCLAW Simulator, developed by Ma et al. (ICML 2023).