Artur Toshev, Nikolaus Adams, Stefan Adami, Fabian Fritz
We lifted 27 functions out of this paper's own repositories and ran 23 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.
| Repository | Role | Ran |
|---|---|---|
| vgsatorras/egnn | canonical | 14 of 15 |
| RobDHess/Steerable-E3-GNN | canonical | 4 of 7 |
| wu375/simple-physics-simulator-pytorch-geometry | canonical | 3 of 3 |
| tumaer/jax-sph | extension | 2 of 2 |
| Function | Status | Where it lives |
|---|---|---|
| adj_bce | Ran | vgsatorras/egnn/losess.py code served (permissive licence) · get_code("e259537bde4ee2f8") |
| build_mlp | Ran | wu375/simple-physics-simulator-pytorch-geometry/train_or_infer.py pointer only (licence: NONE) · get_code("8e6e07db9def5d01") |
| filter_nodes | Ran | vgsatorras/egnn/utils.py code served (permissive licence) · get_code("1f2836f635c3c24b") |
| get_cormorant_features | Ran | RobDHess/Steerable-E3-GNN/qm9/dataset.py code served (permissive licence) · get_code("935a9b7017cf90ff") |
| get_edges | Ran | vgsatorras/egnn/models/egnn_clean/egnn_clean.py code served (permissive licence) · get_code("c2f9c401f5379aac") |
| get_random_walk_noise_for_position_sequence | Ran | wu375/simple-physics-simulator-pytorch-geometry/train_or_infer.py pointer only (licence: NONE) · get_code("c227c9446cea3325") |
| get_velocity_attr | Ran | vgsatorras/egnn/main_nbody.py code served (permissive licence) · get_code("a1ea415560adc94a") |
| graph2networkx | Ran | vgsatorras/egnn/graph.py code served (permissive licence) · get_code("db8fecb91e9bf92b") |
| load_model | Ran | RobDHess/Steerable-E3-GNN/utils.py code served (permissive licence) · get_code("a8149b815c51038e") |
| make_dataloader | Ran | RobDHess/Steerable-E3-GNN/utils.py code served (permissive licence) · get_code("f9e43abe93b5283e") |
| max_n_nodes | Ran | vgsatorras/egnn/ae_datasets/d_creator.py code served (permissive licence) · get_code("cce3235cb632dda2") |
| networkx2graph | Ran | vgsatorras/egnn/graph.py code served (permissive licence) · get_code("4b396ba7acc5f88e") |
| normalize_res | Ran | vgsatorras/egnn/utils.py code served (permissive licence) · get_code("61447f3db462665b") |
| normalizer | Ran | vgsatorras/egnn/models/ae.py code served (permissive licence) · get_code("8e6bbd30a7336168") |
| rho_evol_fn | Ran | tumaer/jax-sph/jax_sph/solver.py code served (permissive licence) · get_code("b2b87cb713d14139") |
| rho_evol_fn_delta_wrapper | Ran | tumaer/jax-sph/jax_sph/solver.py code served (permissive licence) · get_code("3f463624d5a53697") |
| run_epoch | Ran | RobDHess/Steerable-E3-GNN/nbody/train_nbody.py code served (permissive licence) · get_code("f638b3f28da71d6d") |
| sparse2dense | Ran | vgsatorras/egnn/graph.py code served (permissive licence) · get_code("c5d5c0ef14ebb520") |
| time_diff | Ran | wu375/simple-physics-simulator-pytorch-geometry/train_or_infer.py pointer only (licence: NONE) · get_code("33ca4bbea222551b") |
| unsorted_segment_mean | Ran | vgsatorras/egnn/models/egnn_clean/egnn_clean.py code served (permissive licence) · get_code("bc66f19be57720e7") |
| unsorted_segment_sum | Ran | vgsatorras/egnn/models/egnn_clean/egnn_clean.py code served (permissive licence) · get_code("5fb62a78eb65d3d7") |
| unsorted_segment_sum | Ran | vgsatorras/egnn/models/gcl.py code served (permissive licence) · get_code("dfdf6dc1a0720e57") |
| vae_loss | Ran | vgsatorras/egnn/losess.py code served (permissive licence) · get_code("917b57193d100781") |
| BalancedIrreps | Not yet run | RobDHess/Steerable-E3-GNN/models/balanced_irreps.py code served (permissive licence) · get_code("8d6ee0bdca895865") |
| WeightBalancedIrreps | Not yet run | RobDHess/Steerable-E3-GNN/models/balanced_irreps.py code served (permissive licence) · get_code("a5a087121ff0f1fa") |
| plot_coords | Not yet run | vgsatorras/egnn/utils.py code served (permissive licence) · get_code("117ba303e865d2e8") |
| run_epoch | Not yet run | RobDHess/Steerable-E3-GNN/nbody/train_gravity.py code served (permissive licence) · get_code("c18c2f1f9a27b909") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
Machine learning has been successfully applied to grid-based PDE modeling in various scientific applications. However, learned PDE solvers based on Lagrangian particle discretizations, which are the preferred approach to problems with free surfaces or complex physics, remain largely unexplored. We present LagrangeBench, the first benchmarking suite for Lagrangian particle problems, focusing on temporal coarse-graining. In particular, our contribution is: (a) seven new fluid mechanics datasets (four in 2D and three in 3D) generated with the Smoothed Particle Hydrodynamics (SPH) method including the Taylor-Green vortex, lid-driven cavity, reverse Poiseuille flow, and dam break, each of which includes different physics like solid wall interactions or free surface, (b) efficient JAX-based API with various recent training strategies and three neighbor search routines, and (c) JAX implementation of established Graph Neural Networks (GNNs) like GNS and SEGNN with baseline results. Finally, to measure the performance of learned surrogates we go beyond established position errors and introduce physical metrics like kinetic energy MSE and Sinkhorn distance for the particle distribution. Our codebase is available under the URL: https://github.com/tumaer/lagrangebench. * equal contribution 37th Conference on Neural Information Processing Systems (NeurIPS 2023) Track on Datasets and Benchmarks.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2309.16342")
get_code_for_paper("2309.16342")
have("2309.16342")
Connect an agent — have() is free.