{"id":18930413,"url":"https://github.com/aelnouby/relational-networks","last_synced_at":"2025-06-19T04:33:48.890Z","repository":{"id":218143127,"uuid":"127662007","full_name":"aelnouby/Relational-Networks","owner":"aelnouby","description":"Pytorch implementation of \" A simple neural network module for relational reasoning\" paper aka Relational networks for visual reasoning.","archived":false,"fork":false,"pushed_at":"2018-04-26T14:30:19.000Z","size":32,"stargazers_count":9,"open_issues_count":0,"forks_count":0,"subscribers_count":3,"default_branch":"master","last_synced_at":"2025-04-15T17:16:43.450Z","etag":null,"topics":["clevr","pytorch","relational-networks","visual-reasoning"],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"gpl-3.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/aelnouby.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":"2018-04-01T18:54:40.000Z","updated_at":"2024-09-21T11:30:36.000Z","dependencies_parsed_at":null,"dependency_job_id":"1107a45e-15ad-498e-acce-2d99415b9bc3","html_url":"https://github.com/aelnouby/Relational-Networks","commit_stats":null,"previous_names":["aelnouby/relational-networks"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/aelnouby/Relational-Networks","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/aelnouby%2FRelational-Networks","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/aelnouby%2FRelational-Networks/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/aelnouby%2FRelational-Networks/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/aelnouby%2FRelational-Networks/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/aelnouby","download_url":"https://codeload.github.com/aelnouby/Relational-Networks/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/aelnouby%2FRelational-Networks/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":260687810,"owners_count":23046793,"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":["clevr","pytorch","relational-networks","visual-reasoning"],"created_at":"2024-11-08T11:37:35.646Z","updated_at":"2025-06-19T04:33:43.877Z","avatar_url":"https://github.com/aelnouby.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Relational-Networks\nPytorch implementation of \" A simple neural network module for relational reasoning\" paper aka Relational networks for visual reasoning. https://arxiv.org/abs/1706.01427\n\n\u003cimg src='https://writelatex.s3.amazonaws.com/rtdjcknkvwxj/uploads/806/23070795/1.png?X-Amz-Expires=14400\u0026X-Amz-Date=20180426T141042Z\u0026X-Amz-Algorithm=AWS4-HMAC-SHA256\u0026X-Amz-Credential=AKIAJF667VKUK4OW3LCA/20180426/us-east-1/s3/aws4_request\u0026X-Amz-SignedHeaders=host\u0026X-Amz-Signature=f4fc0388dfee0ebc2fb88b46611348a18ae561aa620e2a24ca43949e20d14f93' /\u003e\n\n## Important Note\n\nThis implementation includes only the visual pipeline for CLEVR dataset. Best validation accuracy acheived with this implementation is **72%** compared to **96.8%** reported in the paper. This result was acheived by applying a learning rate schedule that doubles the learning rate every 20 epochs (motivated by warmup in https://arxiv.org/abs/1706.02677). The paper itself does not discuss any schedules used, running with schedules gets **65%** at best.\n\nPull requests and suggestions are welcome to reproduce the results from the paper.\n\n## Training and Valiation Accuracies with warmup\n\n\n\u003cp float=\"left\"\u003e\n  \u003cimg src='https://user-images.githubusercontent.com/8495451/39027855-db99cb10-4421-11e8-9303-9d1b93e1389a.png' width='420' height='420'/\u003e\n  \u003cimg src='https://user-images.githubusercontent.com/8495451/39027856-dbacaa8c-4421-11e8-8903-7dbcf8f26be5.png' width='420' height='420' /\u003e\n\u003c/p\u003e\n\n## Requirements\n\n- [Pytorch](http://pytorch.org/)\n- [Visdom](https://github.com/facebookresearch/visdom)\n- [H5py](https://www.h5py.org/)\n- [opencv-python](https://anaconda.org/conda-forge/opencv)\n\n\n## Usage\n\n### Train\n\n`python3 runtime`\n\n**Arguments**\n\n- `lr` : Learning rate. default: `2.5e-4`\n- `batch_size`: default : `64`\n- `warmup`: A flag to turn on doubling the learning rate every 20 epochs. default: `False`\n- `save_path`: path to checkpoints. Checkpoints are saved for every new best validation accuracy.\n- `vis_screen`: Visdom env name. default: `RelNet`\n\n## Other Implementations (Visual pipeline)\n\n- https://github.com/rosinality/relation-networks-pytorch\n- https://github.com/mesnico/RelationNetworks-CLEVR\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Faelnouby%2Frelational-networks","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Faelnouby%2Frelational-networks","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Faelnouby%2Frelational-networks/lists"}