{"id":13595619,"url":"https://github.com/white127/QA-deep-learning","last_synced_at":"2025-04-09T13:32:51.336Z","repository":{"id":68120747,"uuid":"60221335","full_name":"white127/QA-deep-learning","owner":"white127","description":"tensorflow and theano cnn code for insurance QA(question Answer matching)","archived":false,"fork":false,"pushed_at":"2018-09-07T03:40:34.000Z","size":14452,"stargazers_count":532,"open_issues_count":25,"forks_count":285,"subscribers_count":38,"default_branch":"master","last_synced_at":"2024-08-01T16:41:46.619Z","etag":null,"topics":[],"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/white127.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}},"created_at":"2016-06-02T01:00:34.000Z","updated_at":"2024-07-27T16:26:03.000Z","dependencies_parsed_at":"2023-03-29T20:47:46.007Z","dependency_job_id":null,"html_url":"https://github.com/white127/QA-deep-learning","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/white127%2FQA-deep-learning","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/white127%2FQA-deep-learning/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/white127%2FQA-deep-learning/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/white127%2FQA-deep-learning/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/white127","download_url":"https://codeload.github.com/white127/QA-deep-learning/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":223394647,"owners_count":17138591,"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":"2024-08-01T16:01:53.754Z","updated_at":"2024-11-06T18:31:22.261Z","avatar_url":"https://github.com/white127.png","language":"Python","funding_links":[],"categories":["Python"],"sub_categories":[],"readme":"Insurance-QA deeplearning model\n======\nThis is a repo for Q\u0026A Mathing, includes some deep learning models, such as CNN、RNN.\u003cbr\u003e\n1. CNN. Basic CNN model from 《Applying Deep Learning To Answer Selection: A Study And An Open Task》\u003cbr\u003e\n2. RNN. RNN seems the best model on Insurance-QA dataset.\u003cbr\u003e\n3. SWEM. SWEM is the fastest, and has good effect on other datasets, such as WikiQA ..., but is seems not so good on Insurance-QA dataset. I think that, SWEM is more suitable for Q\u0026Q matching, not Q\u0026A matching.\u003cbr\u003e\n\n\nIt's hard to say which model is the best in other datasets, you have to choose the most suitable model for you.\u003cbr\u003e\u003cbr\u003e\nMore models are on the way, pay attention to the updates.\u003cbr\u003e\n\n## Requirements\n1. tensorflow 1.4.0\u003cbr\u003e\n2. python3.5\u003cbr\u003e\n\n## Performance\nmargin loss version\u003cbr\u003e\n\nModel/Score | Ins_qa_top1_precision | quora_best_prec\n------------ | ------------- | -------------\nCNN | 62% | None\nLSTM+CNN | 68% | None\nSWEM | \u003c55% | None\n\nlogloss version\u003cbr\u003e\n\nModel/Score | Insqa_top1_precision | quora_best_prec\n------------ | ------------- | -------------\nCNN | None | 79.60%\nLSTM+CNN | None | None\nSWEM | \u003c40% | 82.69%\n\n## Running\nChange configuration to your own environment, just like data pathes\u003cbr\u003e\n    \n    vim config.py\n\nData processing\u003cbr\u003e\n   \n    python3 gen.py\n    \nRun CNN model\u003cbr\u003e\n\n    cd ./cnn/tensorflow \u0026\u0026 python3 insqa_train.py\n    \nIt will take few hours(thousands of epoches) to train this model on a single GPU.\u003cbr\u003e\n    \n## Downloads\n1. You can get Insurance-QA data from here https://github.com/shuzi/insuranceQA\u003cbr\u003e\n2. You can get Quora data from here http://qim.ec.quoracdn.net/quora_duplicate_questions.tsv\u003cbr\u003e\n\n## Links\n1. CNN and RNN textual classification repo  https://github.com/white127/TextClassification_CNN_RNN\u003cbr\u003e\n2. 《Applying Deep Learning To Answer Selection: A Study And An Open Task》\u003cbr\u003e\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fwhite127%2FQA-deep-learning","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fwhite127%2FQA-deep-learning","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fwhite127%2FQA-deep-learning/lists"}