{"id":19336811,"url":"https://github.com/blackbbc/ner","last_synced_at":"2026-06-18T11:31:39.000Z","repository":{"id":182633897,"uuid":"149407032","full_name":"blackbbc/NER","owner":"blackbbc","description":"BiLSTM-ELMo-CNN-CRF for CoNLL 2003","archived":false,"fork":false,"pushed_at":"2019-07-15T06:48:45.000Z","size":1075,"stargazers_count":0,"open_issues_count":1,"forks_count":0,"subscribers_count":1,"default_branch":"elmo","last_synced_at":"2025-02-24T08:13:39.448Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":"","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/blackbbc.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}},"created_at":"2018-09-19T07:04:13.000Z","updated_at":"2023-07-20T16:42:28.000Z","dependencies_parsed_at":null,"dependency_job_id":"c0d708f5-a492-42b6-b1b1-33207b5a9d8d","html_url":"https://github.com/blackbbc/NER","commit_stats":null,"previous_names":["blackbbc/ner"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/blackbbc/NER","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/blackbbc%2FNER","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/blackbbc%2FNER/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/blackbbc%2FNER/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/blackbbc%2FNER/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/blackbbc","download_url":"https://codeload.github.com/blackbbc/NER/tar.gz/refs/heads/elmo","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/blackbbc%2FNER/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":34489019,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-06-18T02:00:06.871Z","response_time":128,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"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":"2024-11-10T03:12:30.859Z","updated_at":"2026-06-18T11:31:38.973Z","avatar_url":"https://github.com/blackbbc.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"## ELMo for NER \n\nThis project is a Tensorflow implementation of \"mainstream\" neural tagging scheme based on works of [Deep contextualized word representations, Peters, \net. al., 2018](https://arxiv.org/pdf/1802.05365.pdf). \n\n\n### Requirements\n\n- python 3.6\n- tensorflow 1.10.0\n- numpy 1.14.3\n- gensim 3.6.0\n- tqdm 4.26.0\n\n### Evaluation\n\n| Model  | Dataset    | Test F1 |\n| :----: | :-------:  | :-----: | \n| Peters et. al | CoNLL 2003 | 92.22(+/-0.10)   |\n| Ours          | CoNLL 2003 | 92.23   |\n\n### Prepare Data\n1. Download pre-trained word vector from [http://nlp.stanford.edu/data/glove.6B.zip](http://nlp.stanford.edu/data/glove.6B.zip), unzip `glove.6B.50d.txt` to `resources/pretrained/glove`.\n2. Download pre-trained elmo models [elmo_2x1024_128_2048cnn_1xhighway_weights.hdf5](https://s3-us-west-2.amazonaws.com/allennlp/models/elmo/2x1024_128_2048cnn_1xhighway/elmo_2x1024_128_2048cnn_1xhighway_weights.hdf5) and [elmo_2x1024_128_2048cnn_1xhighway_options.json](https://s3-us-west-2.amazonaws.com/allennlp/models/elmo/2x1024_128_2048cnn_1xhighway/elmo_2x1024_128_2048cnn_1xhighway_options.json), put them in `resources/elmo`.\n\n### Change `glove` to `word2vec` format\nOpen `glove.6B.50d.txt` with your favorite text editor, add `400000 50` to the first line like this:\n\n```\n400000 50\nthe 0.418 0.24968 ...\nof 0.70853 0.57088 ...\n...\n```\n\n### Train\n\n```\npython elmo_train.py\n```\n\n\n### Training Log\n\n| Epoch  | Loss   | Dev F1 | Test F1 |\n| :----: | :----: | :----: | :-----: | \n| 1      | 32237  | 90.81  | 87.77   |\n| 2      | 12320  | 93.21  | 90.16   |\n| 3      | 8823   | 94.19  | 91.75   |\n| 4      | 6900   | 94.80  | 91.74   |\n| 5      | 5821   | 94.30  | 91.03   |\n| 6      | 4996   | 94.92  | 91.26   |\n| 7      | 4467   | 95.18  | 92.06   |\n| 8      | 3869   | 95.06  | 91.54   |\n| 9      | 3483   | 95.13  | 91.88   |\n| 10     | 3500   | 95.42  | 91.66   |\n| 11     | 2989   | 95.01  | 91.82   |\n| 12     | 2770   | 95.39  | 91.70   |\n| 13     | 2649   | 95.39  | 91.68   |\n| 14     | 2529   | 95.44  | **92.23**   |\n| 15     | 2407   | 95.02  | 91.77   |\n| 16     | 2140   | 95.13  | 91.80   |\n| 17     | 2149   | 95.09  | 92.06   |\n| 18     | 1935   | 95.22  | 91.58   |\n| 19     | 1946   | 94.88  | 91.91   |\n| 20     | 1767   | 95.35  | 92.13   |\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fblackbbc%2Fner","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fblackbbc%2Fner","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fblackbbc%2Fner/lists"}