{"id":13752300,"url":"https://github.com/4paradigm/autox","last_synced_at":"2025-05-16T13:06:07.400Z","repository":{"id":37932035,"uuid":"388068949","full_name":"4paradigm/AutoX","owner":"4paradigm","description":"AutoX is an efficient automl tool, which is mainly aimed at data mining tasks with tabular data.","archived":false,"fork":false,"pushed_at":"2023-02-14T14:21:58.000Z","size":75504,"stargazers_count":522,"open_issues_count":20,"forks_count":138,"subscribers_count":13,"default_branch":"master","last_synced_at":"2025-05-12T08:59:52.246Z","etag":null,"topics":["kaggle","machine-learning","python"],"latest_commit_sha":null,"homepage":"https://autox.readthedocs.io","language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/4paradigm.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}},"created_at":"2021-07-21T09:45:41.000Z","updated_at":"2025-05-04T15:52:15.000Z","dependencies_parsed_at":"2023-02-16T16:16:16.073Z","dependency_job_id":null,"html_url":"https://github.com/4paradigm/AutoX","commit_stats":null,"previous_names":[],"tags_count":29,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/4paradigm%2FAutoX","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/4paradigm%2FAutoX/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/4paradigm%2FAutoX/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/4paradigm%2FAutoX/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/4paradigm","download_url":"https://codeload.github.com/4paradigm/AutoX/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":254535827,"owners_count":22087399,"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":["kaggle","machine-learning","python"],"created_at":"2024-08-03T09:01:03.303Z","updated_at":"2025-05-16T13:06:07.382Z","avatar_url":"https://github.com/4paradigm.png","language":"Jupyter Notebook","readme":"[English](./README_EN.md) | 简体中文\n\u003cimg src=\"./img/logo.png\" width = \"1500\" alt=\"logo\" align=center /\u003e\n\n# AutoX是什么？\nAutoX一个高效的自动化机器学习工具。\n它的特点包括:\n- 效果出色: AutoX在多个kaggle数据集上，效果显著优于其他解决方案(见[效果对比](#效果对比))。\n- 简单易用: AutoX的接口和sklearn类似，方便上手使用。\n- 通用: 适用于分类和回归问题。\n- 自动化: 无需人工干预，全自动的数据清洗、特征工程、模型调参等步骤。\n- 灵活性: 各组件解耦合，能单独使用，对于自动机器学习效果不满意的地方，可以结合专家知识，AutoX提供灵活的接口。\n- 比赛上分点总结：整理并公开历史比赛的上分点。\n\n# AutoX包含什么内容\n- [autox_competition](autox/autox_competition/README.md): 主要针对于表格类型的数据挖掘竞赛\n- [autox_server](autox/autox_server/README.md): 用于上线部署的automl服务\n- [autox_interpreter](autox/autox_interpreter/README.md): 机器学习可解释功能\n- [autox_nlp](autox/autox_nlp/README.md): 对文本列进行处理的自动化工具\n- [autox_recommend](autox/autox_recommend/README.md): 推荐系统的自动机器学习\n- [autox_video](autox/autox_video/README.md): 应用于视频分类任务的自动机器学习框架\n\n# 加入社区\n\u003cimg src=\"./img/qr_code_community.png\" width = \"200\" height = \"260\" alt=\"AutoX社区\" align=center /\u003e\n\n# 框架\n## autox_competition\n\u003cimg src=\"./autox/autox_competition/img/framework.png\" alt=\"autox_competition framework\" align=center /\u003e\n\n## autox_recommend\n\u003cimg src=\"./autox/autox_recommend/img/framework_0525.png\" alt=\"autox_recommend framework\" align=center /\u003e\n\n## autox_video\n\u003cimg src=\"./autox/autox_video/resources/framework.png\" alt=\"autox_video framework\" align=center /\u003e\n\n# 如何为AutoX贡献\n[如何为AutoX贡献](./how_to_contribute.md)\n\n# 目录\n\u003c!-- TOC --\u003e\n\n- [AutoX是什么？](#AutoX是什么？)\n- [AutoX包含什么内容](#AutoX包含什么内容)\n- [加入社区](#加入社区)\n- [目录](#目录)\n- [安装](#安装)\n- [如何为AutoX贡献](#如何为AutoX贡献)\n- [快速上手](#快速上手)\n- [效果对比](#效果对比)\n- [TODO](#TODO)\n- [错误排查](#错误排查)\n\n\u003c!-- /TOC --\u003e\n# 安装\n\n### github仓库安装\n```\ngit clone https://github.com/4paradigm/autox.git\npip install ./autox\n```\n\n### pip安装\n```\n## pip安装包可能更新不及时，建议用github安装方式安装最新版本\n!pip install automl-x -i https://www.pypi.org/simple/\n```\n\n# 快速上手\n- [autox打比赛](autox/autox_competition/README.md)\n- [autox上线部署](autox/autox_server/README.md)\n- [autox可解释](autox/autox_interpreter/README.md)\n- [特征工程](autox/autox_competition/feature_engineer/README.md)\n\n# 社区案例\n[汽车销量预测](./demo/汽车销量预测/README.md)\n\n# 比赛案例\n见demo文件夹\n\n数据集下载链接：https://pan.baidu.com/s/1p38OuP8_FJp2P_wJwhdFiw?pwd=8mxf\n# 效果对比\n## 不同任务下的效果提升百分比\n|data_type | 对比AutoGluon | 对比H2o |\n|----- | ------------- | ----------- |\n|binary classification | 20.44% | 2.98% |\n|regression | 37.54% | 39.66% |\n|time-series | 28.40% | 32.46% |\n\n## 详细数据集对比\n|data_type | single-or-multi | data_name | metric | AutoX | AutoGluon | H2o |\n|----- | ------------- | ----------- |---------------- |---------------- | ----------------|----------------|\n|binary classification | single-table | [Springleaf](https://www.kaggle.com/c/springleaf-marketing-response/)  | auc | 0.78865 | 0.61141 | 0.78186 |\n|binary classification-nlp | single-table |[stumbleupon](https://www.kaggle.com/c/stumbleupon/)  | auc | 0.87177 | 0.81025 | 0.79039 |\n|binary classification | single-table |[santander](https://www.kaggle.com/c/santander-customer-transaction-prediction/)  | auc | 0.89196 | 0.64643 | 0.88775 |\n|binary classification | multi-table |[IEEE](https://www.kaggle.com/c/ieee-fraud-detection/)  | accuracy | 0.920809 | 0.724925 | 0.907818 |\n|regression | single-table |[ventilator](https://www.kaggle.com/c/ventilator-pressure-prediction/)  | mae | 0.755 | 8.434 | 4.221 |\n|regression | single-table |[Allstate Claims Severity](https://www.kaggle.com/c/allstate-claims-severity)| mae | 1137.07885 | 1173.35917 | 1163.12014 |\n|regression | single-table |[zhidemai](https://www.automl.ai/competitions/19)   | mse | 1.0034 | 1.9466 | 1.1927|\n|regression | single-table |[Tabular Playground Series - Aug 2021](https://www.kaggle.com/c/tabular-playground-series-aug-2021) | rmse | 7.87731 | 10.3944 | 7.8895|\n|regression | single-table |[House Prices](https://www.kaggle.com/c/house-prices-advanced-regression-techniques/)  | rmse | 0.13043 | 0.13104 | 0.13161 |\n|regression | single-table |[Restaurant Revenue](https://www.kaggle.com/c/restaurant-revenue-prediction/)| rmse | 2133204.32146 | 31913829.59876 | 28958013.69639 |\n|regression | multi-table  |[Elo Merchant Category Recommendation](https://www.kaggle.com/c/elo-merchant-category-recommendation/)| rmse | 3.72228 | 3.80801 | 22.88899 |\n|regression-ts | single-table  |[Demand Forecasting](https://www.kaggle.com/c/demand-forecasting-kernels-only/)| smape | 13.79241 | 25.39182 | 18.89678 |\n|regression-ts | multi-table  |[Walmart Recruiting](https://www.kaggle.com/c/walmart-recruiting-store-sales-forecasting/)| wmae | 4660.99174 | 5024.16179 | 5128.31622 |\n|regression-ts | multi-table  |[Rossmann Store Sales](https://www.kaggle.com/c/rossmann-store-sales/)| RMSPE | 0.13850 | 0.20453 | 0.35757 |\n|regression-cv | single-table |[PetFinder](https://www.kaggle.com/competitions/petfinder-pawpularity-score/overview/)  | rmse | 20.1327 | 23.1732 | 21.0586 |\n\n# AutoX成就\n### 企业支持\n\n### 比赛获奖\n- [2021阿里云基础设施供应链大赛-冠军方案](https://tianchi.aliyun.com/forum/postDetail?postId=344505)\n- [kaggle-H\u0026M个性化推荐-金牌方案](https://www.kaggle.com/competitions/h-and-m-personalized-fashion-recommendations/discussion/324158)\n- [AutoX获得CCF A类会议ACM Multimedia举办的视频分类任务冠军](http://auto-video-captions.top/2022/leaderboard)\n\n\n# TODO\n功能开发完成后，发布相应的使用demo\n- [ ] 多分类任务\n\n若有其他希望AutoX支持的功能，欢迎提issue!\n欢迎填写[用户调研问卷](https://www.wjx.cn/vj/YOwSFHN.aspx)，让AutoX变得更好!\n\n## 错误排查\n|错误信息|解决办法|\n|------|------|\n","funding_links":[],"categories":["其他_机器学习与深度学习"],"sub_categories":[],"project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2F4paradigm%2Fautox","html_url":"https://awesome.ecosyste.ms/projects/github.com%2F4paradigm%2Fautox","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2F4paradigm%2Fautox/lists"}