{"id":13936686,"url":"https://github.com/jimfleming/recurrent-entity-networks","last_synced_at":"2025-07-19T22:31:34.605Z","repository":{"id":143133295,"uuid":"74244129","full_name":"jimfleming/recurrent-entity-networks","owner":"jimfleming","description":"TensorFlow implementation of \"Tracking the World State with Recurrent Entity Networks\".","archived":false,"fork":false,"pushed_at":"2017-11-02T20:31:27.000Z","size":327,"stargazers_count":273,"open_issues_count":2,"forks_count":69,"subscribers_count":17,"default_branch":"master","last_synced_at":"2024-08-08T23:24:16.687Z","etag":null,"topics":["deep-learning","machine-learning","natural-language-processing","recurrent-neural-networks","tensorflow"],"latest_commit_sha":null,"homepage":"https://arxiv.org/abs/1612.03969","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/jimfleming.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":"2016-11-20T00:34:37.000Z","updated_at":"2024-05-26T19:38:26.000Z","dependencies_parsed_at":null,"dependency_job_id":"a4f756dc-2e35-46a3-a430-d5bc0347f834","html_url":"https://github.com/jimfleming/recurrent-entity-networks","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jimfleming%2Frecurrent-entity-networks","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jimfleming%2Frecurrent-entity-networks/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jimfleming%2Frecurrent-entity-networks/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jimfleming%2Frecurrent-entity-networks/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/jimfleming","download_url":"https://codeload.github.com/jimfleming/recurrent-entity-networks/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":226686729,"owners_count":17666928,"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":["deep-learning","machine-learning","natural-language-processing","recurrent-neural-networks","tensorflow"],"created_at":"2024-08-07T23:02:54.699Z","updated_at":"2024-11-27T04:31:19.944Z","avatar_url":"https://github.com/jimfleming.png","language":"Python","funding_links":[],"categories":["Python"],"sub_categories":[],"readme":"# Recurrent Entity Networks\n\nThis repository contains an independent TensorFlow implementation of recurrent entity networks from [Tracking the World State with\nRecurrent Entity Networks](https://arxiv.org/abs/1612.03969). This paper introduces the first method to solve all of the bAbI tasks using 10k training examples. The author's original Torch implementation is now available [here](https://github.com/facebook/MemNN/tree/master/EntNet-babi).\n\n\u003cimg src=\"assets/diagram.png\" alt=\"Diagram of recurrent entity network architecture\" width=\"886\" height=\"658\"\u003e\n\n## Results\n\nPercent error for each task, comparing those in the paper to the implementation contained in this repository.\n\nTask | EntNet (paper) | EntNet (repo)\n--- | --- | ---\n1: 1 supporting fact | 0 | 0\n2: 2 supporting facts | 0.1 | 3.0\n3: 3 supporting facts | 4.1 | ?\n4: 2 argument relations | 0 | 0\n5: 3 argument relations | 0.3 | ?\n6: yes/no questions | 0.2 | 0\n7: counting | 0 | 0\n8: lists/sets | 0.5 | 0\n9: simple negation | 0.1 | 0\n10: indefinite knowledge | 0.6 | 0\n11: basic coreference | 0.3 | 0\n12: conjunction | 0 | 0\n13: compound coreference | 1.3 | 0\n14: time reasoning | 0 | 0\n15: basic deduction | 0 | 0\n16: basic induction | 0.2 | 0\n17: positional reasoning | 0.5 | 1.7\n18: size reasoning | 0.3 | 1.5\n19: path finding | 2.3 | 0\n20: agents motivation | 0 | 0\n**Failed Tasks** | 0 | ?\n**Mean Error** | 0.5 | ?\n\nNOTE: Some of these tasks (16 and 19, in particular) required a change in learning rate schedule to reliably converge.\n\n## Setup\n\n1. Download the datasets by running [download_babi.sh](download_babi.sh) or from [The bAbI Project](https://research.facebook.com/research/babi/).\n2. Run [prep_data.py](entity_networks/prep_data.py) which will convert the datasets into [TFRecords](https://www.tensorflow.org/programmers_guide/reading_data#standard_tensorflow_format).\n3. Run `python -m entity_networks.main` to begin training on QA1.\n\n## Major Dependencies\n\n- TensorFlow v1.1.0\n\n(For additional dependencies see [requirements.txt](requirements.txt))\n\n## Thanks!\n\n- Thanks to Mikael Henaff for providing details about their paper over Thanksgiving break. :)\n- Thanks to Andy Zhang ([@zhangandyx](https://twitter.com/zhangandyx)) for helping me troubleshoot numerical instabilities.\n- Thanks to Mike Young ([@mikalyoung](https://github.com/mikalyoung)) for providing results on some of the longer tasks.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fjimfleming%2Frecurrent-entity-networks","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fjimfleming%2Frecurrent-entity-networks","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fjimfleming%2Frecurrent-entity-networks/lists"}