{"id":18939198,"url":"https://github.com/4paradigm/featinsight","last_synced_at":"2026-03-22T10:30:16.112Z","repository":{"id":214035090,"uuid":"674603270","full_name":"4paradigm/FeatInsight","owner":"4paradigm","description":"FeatInsight is a feature platform based on OpenMLDB","archived":false,"fork":false,"pushed_at":"2024-07-23T11:47:31.000Z","size":3477,"stargazers_count":13,"open_issues_count":26,"forks_count":8,"subscribers_count":10,"default_branch":"main","last_synced_at":"2024-12-31T22:12:20.759Z","etag":null,"topics":["feature","feature-extraction","feature-platform","feature-store"],"latest_commit_sha":null,"homepage":"","language":"Vue","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-zh.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":"2023-08-04T10:46:30.000Z","updated_at":"2024-12-09T09:25:02.000Z","dependencies_parsed_at":"2024-01-22T02:39:02.984Z","dependency_job_id":"e98e6657-2b15-4e5c-86d9-8031b69188d9","html_url":"https://github.com/4paradigm/FeatInsight","commit_stats":null,"previous_names":["4paradigm/feature-platform"],"tags_count":1,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/4paradigm%2FFeatInsight","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/4paradigm%2FFeatInsight/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/4paradigm%2FFeatInsight/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/4paradigm%2FFeatInsight/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/4paradigm","download_url":"https://codeload.github.com/4paradigm/FeatInsight/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":239937701,"owners_count":19721484,"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":["feature","feature-extraction","feature-platform","feature-store"],"created_at":"2024-11-08T12:16:43.322Z","updated_at":"2026-03-22T10:30:16.042Z","avatar_url":"https://github.com/4paradigm.png","language":"Vue","funding_links":[],"categories":[],"sub_categories":[],"readme":"# FeatInsight - 基于 OpenMLDB 的特征平台\n\n## 介绍\n\nFeatInsight 是一个先进的特征计算和存储服务，利用 [OpenMLDB](https://github.com/4paradigm/OpenMLDB) 实现高效的特征计算、管理和编排。FeatInsight 提供简便易用的 UI 界面，用户可以进行机器学习特征开发的全流程，包括数据的导入、查看、编辑，特征的生成、存储、上线等功能。针对离线场景中，用户可以选择特征生成离线样本用于后续的机器学习开发；针对在线场景中，用户可以选择特征创建特征服务，实现实时特征计算。\n\n![](./bigscreen.png)\n\nFeatInsight 的主要目的是解决在机器学习项目中常见的问题，包括简便快捷地进行特征提取、转换、组合、选择以及血缘管理，特征的重用和共享，特征服务版本控制，以及确保在训练和推理过程中使用的特征数据的一致和可靠。应用场景包括 上线在线特征服务，搭建 MLOps工作流，搭建 FeatureStore平台，复用开源特征方案，以及作为机器学习业务组件应用于推荐系统、自然语言处理、金融医疗等领域机器学习落地方案中。\n\nFeatInsight 提供的[主要功能](https://openmldb.ai/docs/zh/main/app_ecosystem/feat_insight/functions/index.html)包括：数据管理，特征管理，上线特征服务，离线样本导出，SQL实验室，预计算特征等。\n\n## 安装部署\n\nFeatInsight 提供多种部署方式，详情请参见[文档](https://openmldb.ai/docs/zh/main/app_ecosystem/feat_insight/install/index.html)\n\n### Docker 镜像\n\n准备配置文件并命名为 `application.yml`。\n\n```\nserver:\n  port: 8888\n \nopenmldb:\n  zk_cluster: 127.0.0.1:2181\n  zk_path: /openmldb\n  apiserver: 127.0.0.1:9080\n```\n\n启动 Docker 容器。\n\n```\ndocker run -d -p 8888:8888 -v `pwd`/application.yml:/app/application.yml registry.cn-shenzhen.aliyuncs.com/tobe43/featinsight\n```\n\n#### All-in-One 镜像\n内置OpenMLDB部署以及配置文件的镜像。\n```\ndocker run -d -p 8888:8888 registry.cn-shenzhen.aliyuncs.com/tobe43/portable-openmldb\n```\n\n### 安装包\n\n准备配置文件`application.yml`。\n```\nwget https://openmldb.ai/download/featinsight/featinsight-0.1.0-SNAPSHOT.jar\n\njava -jar ./featinsight-0.1.0-SNAPSHOT.jar\n```\n\n\n## 使用流程\n\n使用任意网页浏览器访问 FeatInsight 服务地址 http://127.0.0.1:8888/ 。\n\n\nFeatInsight 的大致使用流程如下：\n1. 导入数据：使用 SQL 命令或前端表单进行创建数据库、创建数据表、导入在线数据和导入离线数据等操作。\n2. 创建特征：使用 SQL 语句来定义特征视图，FeatInsight 将使用 SQL 编译器进行特征分析并创建对应的特征。\n3. 离线场景：选择想要导入的特征，可以同时选择不同特征视图的特征，并使用分布式计算把样本文件导入到本地或分布式存储。\n3. 在线场景：选择想要上线的特征，一键发布成在线特征抽取服务，然后可使用 HTTP 客户端进行请求和返回在线特征抽取结果。\n\n我们提供了一个简单的例子来展示如何 FeatInsight 的具体使用流程，请参见[快速入门](https://openmldb.ai/docs/zh/main/app_ecosystem/feat_insight/quickstart.html)。\n\n## 产品文档\n更多内容请参考 [FeatInsight 产品文档](https://openmldb.ai/docs/zh/main/app_ecosystem/feat_insight/index.html)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2F4paradigm%2Ffeatinsight","html_url":"https://awesome.ecosyste.ms/projects/github.com%2F4paradigm%2Ffeatinsight","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2F4paradigm%2Ffeatinsight/lists"}