{"id":13605362,"url":"https://github.com/MachineLearningSystem/marius","last_synced_at":"2025-04-12T05:32:43.495Z","repository":{"id":185461852,"uuid":"496242079","full_name":"MachineLearningSystem/marius","owner":"MachineLearningSystem","description":"Large scale graph learning on a single machine. ","archived":false,"fork":true,"pushed_at":"2022-05-24T21:04:59.000Z","size":5450,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2024-08-02T19:37:20.870Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":"https://marius-project.org","language":null,"has_issues":false,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":"marius-team/marius","license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/MachineLearningSystem.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":"CONTRIBUTING.md","funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null}},"created_at":"2022-05-25T13:26:22.000Z","updated_at":"2022-05-14T11:21:07.000Z","dependencies_parsed_at":null,"dependency_job_id":"e35ee973-7d19-4b13-ad88-176b8003aff6","html_url":"https://github.com/MachineLearningSystem/marius","commit_stats":null,"previous_names":["machinelearningsystem/marius"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/MachineLearningSystem%2Fmarius","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/MachineLearningSystem%2Fmarius/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/MachineLearningSystem%2Fmarius/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/MachineLearningSystem%2Fmarius/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/MachineLearningSystem","download_url":"https://codeload.github.com/MachineLearningSystem/marius/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":223497745,"owners_count":17155199,"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":[],"created_at":"2024-08-01T19:00:57.822Z","updated_at":"2024-11-07T10:30:27.763Z","avatar_url":"https://github.com/MachineLearningSystem.png","language":null,"readme":"# Marius #\n\nMarius is a system for training graph neural networks and embeddings for large-scale graphs on a single machine.\n\nMarius ([OSDI '21 Paper](https://www.usenix.org/conference/osdi21/presentation/mohoney)) is designed to mitigate/reduce data movement overheads using:\n- Pipelined training and IO\n- Partition caching and buffer-aware data orderings\n\nWe scale graph neural network training ([preprint](https://arxiv.org/abs/2202.02365)) through:\n- Optimized datastructures for neighbor sampling and GNN aggregation\n- Out-of-core GNN training \n\n## Build and Install ##\n\n### Requirements ###\n\n* CUDA \u003e= 10.1\n* CuDNN \u003e= 7 \n* pytorch \u003e= 1.8\n* python \u003e= 3.6\n* GCC \u003e= 7 (On Linux) or Clang 12.0 (On MacOS)\n* cmake \u003e= 3.12\n* make \u003e= 3.8\n\n### Pip Installation ###\n\n```\ngit clone https://github.com/marius-team/marius.git\npip3 install .\n```\n\n\n\nThe Python API can be accessed with ``import marius``\n\nThe following commands will be installed:\n- marius_train: Train models using configuration files and the command line\n- marius_eval: Command line model evaluation\n- marius_preprocess: Built-in dataset downloading and preprocessing\n- marius_predict: Batch inference tool for link prediction or node classification\n\n## Command Line Training ##\n\nFirst make sure marius is installed with `pip3 install .` \n\nPreprocess dataset the FB15K_237 dataset with `marius_preprocess --dataset fb15k_237 --output_dir datasets/fb15k_237_example/`\n\nTrain example configuration file (assuming we are in the repo root directory) `marius_train examples/configuration/fb15k_237.yaml`\n\nAfter running this configuration, the MRR output by the system should be about .25 after 10 epochs.\n\nPerform batch inference on the test set with `marius_predict --config examples/configuration/fb15k_237.yaml --metrics mrr --save_scores --save_ranks`\n\nSee the [full example](http://marius-project.org/marius/examples/config/lp_fb15k237.html#small-scale-link-prediction-fb15k-237) for details.\n\n## Python API ##\n\n\nSee the [documentation](http://marius-project.org/marius/examples/python/index.html#) for Python API usage and examples.\n\n\n## Citing Marius ##\nMarius (out-of-core graph embeddings)\n```\n@inproceedings {273733,\n    author = {Jason Mohoney and Roger Waleffe and Henry Xu and Theodoros Rekatsinas and Shivaram Venkataraman},\n    title = {Marius: Learning Massive Graph Embeddings on a Single Machine},\n    booktitle = {15th {USENIX} Symposium on Operating Systems Design and Implementation ({OSDI} 21)},\n    year = {2021},\n    isbn = {978-1-939133-22-9},\n    pages = {533--549},\n    url = {https://www.usenix.org/conference/osdi21/presentation/mohoney},\n    publisher = {{USENIX} Association},\n    month = jul,\n}\n```\n\nMarius++ (out-of-core GNN training)\n```\n@misc{waleffe2022marius,\n  doi = {10.48550/ARXIV.2202.02365},\n  url = {https://arxiv.org/abs/2202.02365},\n  author = {Waleffe, Roger and Mohoney, Jason and Rekatsinas, Theodoros and Venkataraman, Shivaram},\n  keywords = {Machine Learning (cs.LG), Databases (cs.DB), FOS: Computer and information sciences, FOS: Computer and information sciences},\n  title = {Marius++: Large-Scale Training of Graph Neural Networks on a Single Machine},\n  publisher = {arXiv},\n  year = {2022},\n```\n","funding_links":[],"categories":["Paper-Code"],"sub_categories":["GNN"],"project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FMachineLearningSystem%2Fmarius","html_url":"https://awesome.ecosyste.ms/projects/github.com%2FMachineLearningSystem%2Fmarius","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FMachineLearningSystem%2Fmarius/lists"}