{"id":22458885,"url":"https://github.com/s1998/all-but-the-top","last_synced_at":"2025-03-27T14:17:10.343Z","repository":{"id":67610873,"uuid":"211740686","full_name":"s1998/All-but-the-top","owner":"s1998","description":null,"archived":false,"fork":false,"pushed_at":"2019-11-26T15:14:38.000Z","size":82,"stargazers_count":2,"open_issues_count":0,"forks_count":1,"subscribers_count":3,"default_branch":"master","last_synced_at":"2025-02-01T18:27:53.349Z","etag":null,"topics":["abusive-language","davidson","deep-learning","embeddings","f1-acc","hate-speech","iclr","iclr2018","natural-language-processing","word-embeddings"],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/s1998.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"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":"2019-09-29T23:44:14.000Z","updated_at":"2021-06-30T03:18:41.000Z","dependencies_parsed_at":null,"dependency_job_id":"4e40be33-1252-4542-91fe-8d5ba356441d","html_url":"https://github.com/s1998/All-but-the-top","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/s1998%2FAll-but-the-top","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/s1998%2FAll-but-the-top/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/s1998%2FAll-but-the-top/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/s1998%2FAll-but-the-top/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/s1998","download_url":"https://codeload.github.com/s1998/All-but-the-top/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":245858881,"owners_count":20684062,"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":["abusive-language","davidson","deep-learning","embeddings","f1-acc","hate-speech","iclr","iclr2018","natural-language-processing","word-embeddings"],"created_at":"2024-12-06T08:42:08.738Z","updated_at":"2025-03-27T14:17:10.322Z","avatar_url":"https://github.com/s1998.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"\n# All-but-the-top\n\nImplementation of the paper [All-but-the-top](https://openreview.net/forum?id=HkuGJ3kCb) from ICLR 2018.\n\n## Instructions to use\n\nTo run, use the file runner.py\n\nLibraries used: Keras with tensorflow backend.\n\n### Sample plot\n\n![image-of-principal-components](https://raw.githubusercontent.com/s1998/All-but-the-top/master/images/gloveFreqPLot.png)\n\nTwo principal components of the embeddings with color map for frequency.\n\n\n## Results \n\nResults obtained on Davidson et al (2017) using two different embeddings.\n\n\u003ctable\u003e\n  \u003ctr\u003e\n  \t\u003ctd\u003e Model \u003c/td\u003e\n  \t\u003ctd colspan=\"4\"\u003e Preprocessed \u003c/td\u003e\n  \t\u003ctd colspan=\"4\"\u003e Postprocessed \u003c/td\u003e\n  \u003c/tr\u003e\n  \u003ctr\u003e\n  \t\u003ctd\u003e\u003c/td\u003e \n  \t\u003ctd\u003e  P  \u003c/td\u003e\n  \t\u003ctd\u003e  R  \u003c/td\u003e\n  \t\u003ctd\u003e  F1  \u003c/td\u003e \n  \t\u003ctd\u003e  Acc  \u003c/td\u003e \n  \t\u003ctd\u003e  P  \u003c/td\u003e \n  \t\u003ctd\u003e  R  \u003c/td\u003e \n  \t\u003ctd\u003e  F1  \u003c/td\u003e \n  \t\u003ctd\u003e  Acc  \u003c/td\u003e \n  \u003c/tr\u003e\n  \u003ctr\u003e\n  \t\u003ctd\u003eAvgPool \u003c/td\u003e \n  \t\u003ctd\u003e 0.811  \u003c/td\u003e\n  \t\u003ctd\u003e 0.721  \u003c/td\u003e\n  \t\u003ctd\u003e 0.756  \u003c/td\u003e \n  \t\u003ctd\u003e 0.762  \u003c/td\u003e \n  \t\u003ctd\u003e 0.855  \u003c/td\u003e \n  \t\u003ctd\u003e 0.887  \u003c/td\u003e \n  \t\u003ctd\u003e 0.862  \u003c/td\u003e \n  \t\u003ctd\u003e 0.887  \u003c/td\u003e \n  \u003c/tr\u003e\n  \u003ctr\u003e\n  \t\u003ctd\u003eMaxPool \u003c/td\u003e \n  \t\u003ctd\u003e 0.779  \u003c/td\u003e\n  \t\u003ctd\u003e 0.834  \u003c/td\u003e\n  \t\u003ctd\u003e 0.792  \u003c/td\u003e \n  \t\u003ctd\u003e 0.787  \u003c/td\u003e \n  \t\u003ctd\u003e 0.888  \u003c/td\u003e \n  \t\u003ctd\u003e 0.903  \u003c/td\u003e \n  \t\u003ctd\u003e 0.884  \u003c/td\u003e \n  \t\u003ctd\u003e 0.903  \u003c/td\u003e \n  \u003c/tr\u003e\n  \u003ctr\u003e\n  \t\u003ctd\u003eCNN     \u003c/td\u003e \n  \t\u003ctd\u003e 0.885  \u003c/td\u003e\n  \t\u003ctd\u003e 0.903  \u003c/td\u003e\n  \t\u003ctd\u003e 0.880  \u003c/td\u003e \n  \t\u003ctd\u003e 0.903  \u003c/td\u003e \n  \t\u003ctd\u003e 0.890  \u003c/td\u003e \n  \t\u003ctd\u003e 0.905  \u003c/td\u003e \n  \t\u003ctd\u003e 0.892  \u003c/td\u003e \n  \t\u003ctd\u003e 0.905  \u003c/td\u003e \n  \u003c/tr\u003e\n  \u003ctr\u003e\n  \t\u003ctd\u003eGRU     \u003c/td\u003e \n  \t\u003ctd\u003e 0.894  \u003c/td\u003e\n  \t\u003ctd\u003e 0.907  \u003c/td\u003e\n  \t\u003ctd\u003e 0.898  \u003c/td\u003e \n  \t\u003ctd\u003e 0.907  \u003c/td\u003e \n  \t\u003ctd\u003e 0.899  \u003c/td\u003e \n  \t\u003ctd\u003e 0.914  \u003c/td\u003e \n  \t\u003ctd\u003e 0.902  \u003c/td\u003e \n  \t\u003ctd\u003e 0.914  \u003c/td\u003e \n  \u003c/tr\u003e\n\u003c/table\u003e\n\nEffects of using post processing on Glove Embeddings on Davidson et Al(2017)\n\n\n\u003ctable\u003e\n  \u003ctr\u003e\n  \t\u003ctd\u003e Model \u003c/td\u003e\n  \t\u003ctd colspan=\"4\"\u003e Preprocessed \u003c/td\u003e\n  \t\u003ctd colspan=\"4\"\u003e Postprocessed \u003c/td\u003e\n  \u003c/tr\u003e\n  \u003ctr\u003e\n  \t\u003ctd\u003e\u003c/td\u003e \n  \t\u003ctd\u003e  P  \u003c/td\u003e\n  \t\u003ctd\u003e  R  \u003c/td\u003e\n  \t\u003ctd\u003e  F1  \u003c/td\u003e \n  \t\u003ctd\u003e  Acc  \u003c/td\u003e \n  \t\u003ctd\u003e  P  \u003c/td\u003e \n  \t\u003ctd\u003e  R  \u003c/td\u003e \n  \t\u003ctd\u003e  F1  \u003c/td\u003e \n  \t\u003ctd\u003e  Acc  \u003c/td\u003e \n  \u003c/tr\u003e\n  \u003ctr\u003e\n  \t\u003ctd\u003e AvgPool \u003c/td\u003e \n  \t\u003ctd\u003e 0.787 \u003c/td\u003e\n  \t\u003ctd\u003e 0.732 \u003c/td\u003e\n  \t\u003ctd\u003e 0.754 \u003c/td\u003e\n  \t\u003ctd\u003e 0.782 \u003c/td\u003e\n  \t\u003ctd\u003e 0.898 \u003c/td\u003e\n  \t\u003ctd\u003e 0.893 \u003c/td\u003e\n  \t\u003ctd\u003e 0.868 \u003c/td\u003e\n  \t\u003ctd\u003e 0.883 \u003c/td\u003e\n  \u003c/tr\u003e\n  \u003ctr\u003e\n  \t\u003ctd\u003e MaxPool \u003c/td\u003e \n  \t\u003ctd\u003e 0.703 \u003c/td\u003e\n  \t\u003ctd\u003e 0.756 \u003c/td\u003e\n  \t\u003ctd\u003e 0.724 \u003c/td\u003e\n  \t\u003ctd\u003e 0.751 \u003c/td\u003e\n  \t\u003ctd\u003e 0.891 \u003c/td\u003e\n  \t\u003ctd\u003e 0.887 \u003c/td\u003e\n\t\u003ctd\u003e 0.872 \u003c/td\u003e\n\t\u003ctd\u003e 0.887 \u003c/td\u003e\n  \u003c/tr\u003e\n  \u003ctr\u003e\n  \t\u003ctd\u003e CNN     \u003c/td\u003e \n  \t\u003ctd\u003e 0.838 \u003c/td\u003e\n  \t\u003ctd\u003e 0.887 \u003c/td\u003e\n  \t\u003ctd\u003e 0.861 \u003c/td\u003e\n  \t\u003ctd\u003e 0.887 \u003c/td\u003e\n  \t\u003ctd\u003e 0.875 \u003c/td\u003e\n  \t\u003ctd\u003e 0.893 \u003c/td\u003e\n\t\u003ctd\u003e 0.875 \u003c/td\u003e\n\t\u003ctd\u003e 0.891\u003c/td\u003e\n  \u003c/tr\u003e\n  \u003ctr\u003e\n  \t\u003ctd\u003e GRU     \u003c/td\u003e \n  \t\u003ctd\u003e 0.854 \u003c/td\u003e\n  \t\u003ctd\u003e 0.904 \u003c/td\u003e\n  \t\u003ctd\u003e 0.878 \u003c/td\u003e\n  \t\u003ctd\u003e 0.904 \u003c/td\u003e\n  \t\u003ctd\u003e 0.910 \u003c/td\u003e\n  \t\u003ctd\u003e 0.903 \u003c/td\u003e\n\t\u003ctd\u003e 0.881 \u003c/td\u003e\n\t\u003ctd\u003e 0.903\u003c/td\u003e\n  \u003c/tr\u003e\n\u003c/table\u003e\n\nEffects of using post processing on Word2Vec Embeddings on Davidson et Al(2017)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fs1998%2Fall-but-the-top","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fs1998%2Fall-but-the-top","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fs1998%2Fall-but-the-top/lists"}