{"id":38781239,"url":"https://github.com/atomicarchitects/relaxede3nn","last_synced_at":"2026-01-17T12:25:01.167Z","repository":{"id":279689863,"uuid":"885051530","full_name":"atomicarchitects/RelaxedE3NN","owner":"atomicarchitects","description":"[GRaM at ICML'24] Relaxed Equivariant Graph Neural Networks","archived":false,"fork":false,"pushed_at":"2024-11-08T04:14:40.000Z","size":1120,"stargazers_count":4,"open_issues_count":0,"forks_count":1,"subscribers_count":3,"default_branch":"main","last_synced_at":"2025-12-20T23:45:25.737Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/atomicarchitects.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2024-11-07T21:38:11.000Z","updated_at":"2025-03-17T12:23:01.000Z","dependencies_parsed_at":"2025-02-26T22:33:01.564Z","dependency_job_id":null,"html_url":"https://github.com/atomicarchitects/RelaxedE3NN","commit_stats":null,"previous_names":["atomicarchitects/relaxede3nn"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/atomicarchitects/RelaxedE3NN","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/atomicarchitects%2FRelaxedE3NN","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/atomicarchitects%2FRelaxedE3NN/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/atomicarchitects%2FRelaxedE3NN/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/atomicarchitects%2FRelaxedE3NN/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/atomicarchitects","download_url":"https://codeload.github.com/atomicarchitects/RelaxedE3NN/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/atomicarchitects%2FRelaxedE3NN/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":28508464,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-01-17T11:50:55.898Z","status":"ssl_error","status_checked_at":"2026-01-17T11:50:55.569Z","response_time":85,"last_error":"SSL_connect returned=1 errno=0 peeraddr=140.82.121.6:443 state=error: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2026-01-17T12:25:01.075Z","updated_at":"2026-01-17T12:25:01.151Z","avatar_url":"https://github.com/atomicarchitects.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# RelaxedE3NN\n[GRaM at ICML'24] Relaxed Equivariant Graph Neural Networks.\n\nThis repo contains code for Relaxed Graph Equivariant Neural Networks (https://arxiv.org/abs/2407.20471). \n\n## Install\n\n### Dependencies\n\n#### PyTorch\n\ne3nn requires PyTorch \u003e=1.8.0 For installation instructions, please see the [PyTorch homepage](https://pytorch.org/).\n\n#### torch_geometric\n\nFirst you have to install [pytorch_geometric](https://github.com/rusty1s/pytorch_geometric). For `torch` 1.11 and no CUDA support:\n\n```bash\nCUDA=cpu\n\npip install --upgrade --force-reinstall torch-scatter -f https://data.pyg.org/whl/torch-1.11.0+${CUDA}.html\npip install --upgrade --force-reinstall torch-sparse -f https://data.pyg.org/whl/torch-1.11.0+${CUDA}.html\npip install torch-geometric\n```\n\nSee [here](https://github.com/rusty1s/pytorch_geometric#installation) to get cuda support or newer versions.\n\n#### e3nn\n\n#### Stable (PyPI)\n\n```bash\n$ pip install e3nn\n```\n\n## Organization\n\nWe provide code for two toy experiments. \n### Shape Deformations\nshape_deformations_3d.ipynb contains symmetry breaking examples deforming a cube into a rectangular prism/an asymmetric shape. We demonstrate that the relaxed weights are interpretable through plotting their spherical harmonic projections (see the paper for more detail).\n### Electric Field Simulation\nelectric_field_sim.ipynb contains an example learning the direction of the electric and magnetic force for a charged particle in a magnetic field.\n### Models\nrelaxed_e3nn_conv.py contains a simple relaxed e3nn convolution model made of stacked RelaxedConvolutions. The electric_field_model folder contains an example of how to incorporate the relaxed e3nn layer into a more complicated model based on the sample networks contained in [e3nn](https://docs.e3nn.org/en/latest/). The files modified to contain the relaxed equivariant layer are specifically based SimpleNetwork and NetworkForAGraphWithAttributes in e3nn [here](https://github.com/e3nn/e3nn/blob/main/e3nn/nn/models/v2103/gate_points_networks.py). Within the electric_field_model, relaxed_points_conv.py modifies the [points convolution](https://github.com/e3nn/e3nn/blob/main/e3nn/nn/models/v2103/points_convolution.py) in e3nn. gate_points_message_passing_relaxed.py modifies the [message passing](https://github.com/e3nn/e3nn/blob/main/e3nn/nn/models/v2103/gate_points_message_passing.py) to use the relaxed convolution. gate_points_networks_relaxed.py modifies the sample message passing graph neural network models in e3nn to use the relaxed convolution layer.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fatomicarchitects%2Frelaxede3nn","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fatomicarchitects%2Frelaxede3nn","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fatomicarchitects%2Frelaxede3nn/lists"}