{"id":21297243,"url":"https://github.com/minkaixu/egno","last_synced_at":"2025-07-11T18:32:26.377Z","repository":{"id":220934486,"uuid":"752973633","full_name":"MinkaiXu/EGNO","owner":"MinkaiXu","description":"ICML2024: Equivariant Graph Neural Operator for Modeling 3D Dynamics","archived":false,"fork":false,"pushed_at":"2024-03-27T20:02:10.000Z","size":10775,"stargazers_count":55,"open_issues_count":2,"forks_count":8,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-04-06T07:09:53.587Z","etag":null,"topics":["dynamic-simulations","geometric-deep-learning","graph-neural-networks","molecular-dynamics","neural-operator"],"latest_commit_sha":null,"homepage":"","language":"Python","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/MinkaiXu.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":"2024-02-05T08:13:16.000Z","updated_at":"2025-03-29T11:39:23.000Z","dependencies_parsed_at":"2024-11-28T04:45:40.563Z","dependency_job_id":null,"html_url":"https://github.com/MinkaiXu/EGNO","commit_stats":null,"previous_names":["minkaixu/egno"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/MinkaiXu/EGNO","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/MinkaiXu%2FEGNO","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/MinkaiXu%2FEGNO/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/MinkaiXu%2FEGNO/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/MinkaiXu%2FEGNO/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/MinkaiXu","download_url":"https://codeload.github.com/MinkaiXu/EGNO/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/MinkaiXu%2FEGNO/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":264870549,"owners_count":23676261,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","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":["dynamic-simulations","geometric-deep-learning","graph-neural-networks","molecular-dynamics","neural-operator"],"created_at":"2024-11-21T14:34:29.886Z","updated_at":"2025-07-11T18:32:21.618Z","avatar_url":"https://github.com/MinkaiXu.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Equivariant Graph Neural Operator for Modeling 3D Dynamics\n\n\n[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://github.com/MinkaiXu/EGNO/blob/main/LICENSE)\n[![ArXiv](https://img.shields.io/static/v1?\u0026logo=arxiv\u0026label=Paper\u0026message=Arxiv:EGNO\u0026color=B31B1B)](https://arxiv.org/abs/2401.11037)\n\nThis repository contains the official implementation of our [Equivariant Graph Neural Operator](https://arxiv.org/abs/2401.11037).\n\n\u003e [**Equivariant Graph Neural Operator for Modeling 3D Dynamics**](https://arxiv.org/abs/2401.11037)   \n\u003e Minkai Xu*, Jiaqi Han*, Aaron Lou, Jean Kossaifi, Arvind Ramanathan, Kamyar Azizzadenesheli, Jure Leskovec, Stefano Ermon, Anima Anandkumar\n\u003e \u003cbr\u003e**Stanford University, NVIDIA, Argonne National Laboratory, California Institute of Technology**\u003cbr\u003e\n\n![Cover](assets/overview.png)\n\n## Environment\n\nYou can install the exact environment with `env.yml`:\n\n```bash\nconda env create -f env.yml\n```\n\nor manually install the following packages:\n\n```\npython=3.8.17\npytorch=2.0.1\nscipy=1.10.1\n```\nYou may also need `mdanalysis` if you want to process the protein MD data.\n\n\n## Data Preparation\n\n**1. Simulation dataset**\n\nWe provide the data preprocessing code in `simulation/dataset`. One can simply run\n\n```python\ncd simulation/dataset\npython -u generate_dataset.py\n```\n\n**2. Motion capture dataset**\n\nWe provide our pre-processed dataset as well as the splits in `motion/dataset` folder.\n\n**3. MD17 dataset**\n\nWe provide the splits in `md17` folder. The dataset can be downloaded from [here](http://www.sgdml.org/#datasets) and then placed in `md17` folder.\n\n**4. Protein MD**\n\nWe provide the data preprocessing code in `mdanalysis/preprocess.py`. One can simply run\n\n```python\npython mdanalysis/preprocess.py\n```\n\nafter setting the correct data path specified as the variable `tmp_path` in `preprocess.py`.\n\n\n## Train the EGNO\n\n**1. Simulation dataset**\n\n```bash\npython -u main_simulation_simple_no.py --config_by_file --outf $log_dir\n```\n\n**2. Motion capture**\n\n```bash\npython -u main_mocap_no.py --config_by_file --outf $log_dir\n```\n\n**3. MD17**\n\n```bash\npython -u main_md17_no.py --config_by_file --outf $log_dir\n```\n\n**4. Protein MD**\n\n```bash\npython -u main_mdanalysis_no.py --config_by_file --outf $log_dir\n```\n\n\n## Evaluation\n\nAll evaluations (validation and testing) are conducted along with training.\n\n## Citation\n\nPlease consider citing the our paper if you find it helpful. Thank you!\n\n```\n@article{xu2024equivariant,\n  title={Equivariant Graph Neural Operator for Modeling 3D Dynamics},\n  author={Xu, Minkai and Han, Jiaqi and Lou, Aaron and Kossaifi, Jean and Ramanathan, Arvind and Azizzadenesheli, Kamyar and Leskovec, Jure and Ermon, Stefano and Anandkumar, Anima},\n  journal={arXiv preprint arXiv:2401.11037},\n  year={2024}\n}\n```\n\n## Contact\n\nIf you have any question, welcome to contact me at:\n\nMinkai Xu: minkai@cs.stanford.edu\n\n## Acknowledgement\n\nThis repo is built upon several great codebases, including [EGNN](https://github.com/vgsatorras/egnn) and [GMN](https://github.com/hanjq17/GMN). We thank the authors for their great work and open-sourcing the code!\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fminkaixu%2Fegno","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fminkaixu%2Fegno","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fminkaixu%2Fegno/lists"}