[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"similar-featurestoreorg--serverless-ml-course":3,"tool-featurestoreorg--serverless-ml-course":64},[4,17,27,35,43,56],{"id":5,"name":6,"github_repo":7,"description_zh":8,"stars":9,"difficulty_score":10,"last_commit_at":11,"category_tags":12,"status":16},3808,"stable-diffusion-webui","AUTOMATIC1111\u002Fstable-diffusion-webui","stable-diffusion-webui 是一个基于 Gradio 构建的网页版操作界面，旨在让用户能够轻松地在本地运行和使用强大的 Stable Diffusion 图像生成模型。它解决了原始模型依赖命令行、操作门槛高且功能分散的痛点，将复杂的 AI 绘图流程整合进一个直观易用的图形化平台。\n\n无论是希望快速上手的普通创作者、需要精细控制画面细节的设计师，还是想要深入探索模型潜力的开发者与研究人员，都能从中获益。其核心亮点在于极高的功能丰富度：不仅支持文生图、图生图、局部重绘（Inpainting）和外绘（Outpainting）等基础模式，还独创了注意力机制调整、提示词矩阵、负向提示词以及“高清修复”等高级功能。此外，它内置了 GFPGAN 和 CodeFormer 等人脸修复工具，支持多种神经网络放大算法，并允许用户通过插件系统无限扩展能力。即使是显存有限的设备，stable-diffusion-webui 也提供了相应的优化选项，让高质量的 AI 艺术创作变得触手可及。",162132,3,"2026-04-05T11:01:52",[13,14,15],"开发框架","图像","Agent","ready",{"id":18,"name":19,"github_repo":20,"description_zh":21,"stars":22,"difficulty_score":23,"last_commit_at":24,"category_tags":25,"status":16},1381,"everything-claude-code","affaan-m\u002Feverything-claude-code","everything-claude-code 是一套专为 AI 编程助手（如 Claude Code、Codex、Cursor 等）打造的高性能优化系统。它不仅仅是一组配置文件，而是一个经过长期实战打磨的完整框架，旨在解决 AI 代理在实际开发中面临的效率低下、记忆丢失、安全隐患及缺乏持续学习能力等核心痛点。\n\n通过引入技能模块化、直觉增强、记忆持久化机制以及内置的安全扫描功能，everything-claude-code 能显著提升 AI 在复杂任务中的表现，帮助开发者构建更稳定、更智能的生产级 AI 代理。其独特的“研究优先”开发理念和针对 Token 消耗的优化策略，使得模型响应更快、成本更低，同时有效防御潜在的攻击向量。\n\n这套工具特别适合软件开发者、AI 研究人员以及希望深度定制 AI 工作流的技术团队使用。无论您是在构建大型代码库，还是需要 AI 协助进行安全审计与自动化测试，everything-claude-code 都能提供强大的底层支持。作为一个曾荣获 Anthropic 黑客大奖的开源项目，它融合了多语言支持与丰富的实战钩子（hooks），让 AI 真正成长为懂上",140436,2,"2026-04-05T23:32:43",[13,15,26],"语言模型",{"id":28,"name":29,"github_repo":30,"description_zh":31,"stars":32,"difficulty_score":23,"last_commit_at":33,"category_tags":34,"status":16},2271,"ComfyUI","Comfy-Org\u002FComfyUI","ComfyUI 是一款功能强大且高度模块化的视觉 AI 引擎，专为设计和执行复杂的 Stable Diffusion 图像生成流程而打造。它摒弃了传统的代码编写模式，采用直观的节点式流程图界面，让用户通过连接不同的功能模块即可构建个性化的生成管线。\n\n这一设计巧妙解决了高级 AI 绘图工作流配置复杂、灵活性不足的痛点。用户无需具备编程背景，也能自由组合模型、调整参数并实时预览效果，轻松实现从基础文生图到多步骤高清修复等各类复杂任务。ComfyUI 拥有极佳的兼容性，不仅支持 Windows、macOS 和 Linux 全平台，还广泛适配 NVIDIA、AMD、Intel 及苹果 Silicon 等多种硬件架构，并率先支持 SDXL、Flux、SD3 等前沿模型。\n\n无论是希望深入探索算法潜力的研究人员和开发者，还是追求极致创作自由度的设计师与资深 AI 绘画爱好者，ComfyUI 都能提供强大的支持。其独特的模块化架构允许社区不断扩展新功能，使其成为当前最灵活、生态最丰富的开源扩散模型工具之一，帮助用户将创意高效转化为现实。",107662,"2026-04-03T11:11:01",[13,14,15],{"id":36,"name":37,"github_repo":38,"description_zh":39,"stars":40,"difficulty_score":23,"last_commit_at":41,"category_tags":42,"status":16},3704,"NextChat","ChatGPTNextWeb\u002FNextChat","NextChat 是一款轻量且极速的 AI 助手，旨在为用户提供流畅、跨平台的大模型交互体验。它完美解决了用户在多设备间切换时难以保持对话连续性，以及面对众多 AI 模型不知如何统一管理的痛点。无论是日常办公、学习辅助还是创意激发，NextChat 都能让用户随时随地通过网页、iOS、Android、Windows、MacOS 或 Linux 端无缝接入智能服务。\n\n这款工具非常适合普通用户、学生、职场人士以及需要私有化部署的企业团队使用。对于开发者而言，它也提供了便捷的自托管方案，支持一键部署到 Vercel 或 Zeabur 等平台。\n\nNextChat 的核心亮点在于其广泛的模型兼容性，原生支持 Claude、DeepSeek、GPT-4 及 Gemini Pro 等主流大模型，让用户在一个界面即可自由切换不同 AI 能力。此外，它还率先支持 MCP（Model Context Protocol）协议，增强了上下文处理能力。针对企业用户，NextChat 提供专业版解决方案，具备品牌定制、细粒度权限控制、内部知识库整合及安全审计等功能，满足公司对数据隐私和个性化管理的高标准要求。",87618,"2026-04-05T07:20:52",[13,26],{"id":44,"name":45,"github_repo":46,"description_zh":47,"stars":48,"difficulty_score":23,"last_commit_at":49,"category_tags":50,"status":16},2268,"ML-For-Beginners","microsoft\u002FML-For-Beginners","ML-For-Beginners 是由微软推出的一套系统化机器学习入门课程，旨在帮助零基础用户轻松掌握经典机器学习知识。这套课程将学习路径规划为 12 周，包含 26 节精炼课程和 52 道配套测验，内容涵盖从基础概念到实际应用的完整流程，有效解决了初学者面对庞大知识体系时无从下手、缺乏结构化指导的痛点。\n\n无论是希望转型的开发者、需要补充算法背景的研究人员，还是对人工智能充满好奇的普通爱好者，都能从中受益。课程不仅提供了清晰的理论讲解，还强调动手实践，让用户在循序渐进中建立扎实的技能基础。其独特的亮点在于强大的多语言支持，通过自动化机制提供了包括简体中文在内的 50 多种语言版本，极大地降低了全球不同背景用户的学习门槛。此外，项目采用开源协作模式，社区活跃且内容持续更新，确保学习者能获取前沿且准确的技术资讯。如果你正寻找一条清晰、友好且专业的机器学习入门之路，ML-For-Beginners 将是理想的起点。",84991,"2026-04-05T10:45:23",[14,51,52,53,15,54,26,13,55],"数据工具","视频","插件","其他","音频",{"id":57,"name":58,"github_repo":59,"description_zh":60,"stars":61,"difficulty_score":10,"last_commit_at":62,"category_tags":63,"status":16},3128,"ragflow","infiniflow\u002Fragflow","RAGFlow 是一款领先的开源检索增强生成（RAG）引擎，旨在为大语言模型构建更精准、可靠的上下文层。它巧妙地将前沿的 RAG 技术与智能体（Agent）能力相结合，不仅支持从各类文档中高效提取知识，还能让模型基于这些知识进行逻辑推理和任务执行。\n\n在大模型应用中，幻觉问题和知识滞后是常见痛点。RAGFlow 通过深度解析复杂文档结构（如表格、图表及混合排版），显著提升了信息检索的准确度，从而有效减少模型“胡编乱造”的现象，确保回答既有据可依又具备时效性。其内置的智能体机制更进一步，使系统不仅能回答问题，还能自主规划步骤解决复杂问题。\n\n这款工具特别适合开发者、企业技术团队以及 AI 研究人员使用。无论是希望快速搭建私有知识库问答系统，还是致力于探索大模型在垂直领域落地的创新者，都能从中受益。RAGFlow 提供了可视化的工作流编排界面和灵活的 API 接口，既降低了非算法背景用户的上手门槛，也满足了专业开发者对系统深度定制的需求。作为基于 Apache 2.0 协议开源的项目，它正成为连接通用大模型与行业专有知识之间的重要桥梁。",77062,"2026-04-04T04:44:48",[15,14,13,26,54],{"id":65,"github_repo":66,"name":67,"description_en":68,"description_zh":69,"ai_summary_zh":69,"readme_en":70,"readme_zh":71,"quickstart_zh":72,"use_case_zh":73,"hero_image_url":74,"owner_login":75,"owner_name":76,"owner_avatar_url":77,"owner_bio":78,"owner_company":79,"owner_location":79,"owner_email":80,"owner_twitter":79,"owner_website":81,"owner_url":82,"languages":83,"stars":96,"forks":97,"last_commit_at":98,"license":99,"difficulty_score":23,"env_os":100,"env_gpu":100,"env_ram":100,"env_deps":101,"category_tags":110,"github_topics":111,"view_count":23,"oss_zip_url":79,"oss_zip_packed_at":79,"status":16,"created_at":120,"updated_at":121,"faqs":122,"releases":123},2732,"featurestoreorg\u002Fserverless-ml-course","serverless-ml-course","Serverless Machine Learning Course for building AI-enabled Prediction Services from models and features","serverless-ml-course 是一套专为构建 AI 预测服务设计的开源实战课程，旨在帮助开发者无需精通 Kubernetes 或复杂的云运维知识，即可轻松搭建端到端的机器学习系统。它主要解决了传统机器学习部署中环境配置繁琐、基础设施管理困难以及从模型训练到服务上线门槛过高的问题。\n\n这套课程非常适合具备 Python 基础的数据科学家、机器学习工程师以及希望将模型转化为实际应用的开发者。通过循序渐进的模块设计，学习者只需编写 Python 代码来定义数据管道，底层的特征存储、模型注册及服务调度均由系统自动托管，真正实现了“无服务器”体验。\n\n其独特的技术亮点在于强调“超越笔记本（Beyond Notebooks）”的工程化思维，引导用户利用 Pandas 构建批处理与实时预测流水线，并涵盖从信用卡欺诈检测等真实案例到自定义 UI 界面开发的全流程。课程提供了丰富的视频讲座、幻灯片及动手实验，让使用者能专注于算法逻辑与业务价值，而非被底层架构所困扰，是入门现代化 MLOps 实践的理想指南。","\n![readme header](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ffeaturestoreorg_serverless-ml-course_readme_c0a46b7768d1.jpg)\n\n\n\u003Ca href=\"https:\u002F\u002Fjoin.slack.com\u002Ft\u002Ffeaturestoreorg\u002Fshared_invite\u002Fzt-ssh8dec1-IsvwLdTsRMssjT4~Ru2RKg\" alt=\"slack\"> \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FJoin Slack-blue.svg?logo=slack\" \u002F>\u003C\u002Fa> \n\u003Ca href=\"https:\u002F\u002Fwww.youtube.com\u002Fchannel\u002FUC-LrK8ei6w57RmKeswkU23Q\" alt=\"youtube\"> \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FYoutube-red.svg?logo=Youtube\" \u002F>\u003C\u002Fa>\n\u003Ca href=\"https:\u002F\u002Fdiscord.com\u002Fchannels\u002F1065622064800735282\u002F1077878719881945109\" alt=\"discord\"> \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FJoin discord-white.svg?logo=discord\" \u002F>\u003C\u002Fa> \n\n\n\n# **[Beyond Notebooks - Serverless Machine Learning](https:\u002F\u002Fwww.serverless-ml.org)**\n***Build Batch and Real-Time Prediction Services with Python***\n\n![serverless architecture](\u002Fassets\u002Fimages\u002Fserverless-ml-architecture.svg \"Serverless Architecture\")\n\n# **Overview**\nYou should not need to be an expert in Kubernetes or cloud computing to build an end-to-end service that makes intelligent decisions with the help of a ML model. Serverless Machine Learning (ML) makes it easy to build a system that uses ML models to make predictions. \n\nWith Serverless ML, you do not need to install, upgrade, or operate any systems. You only need to be able to write Python programs that can be scheduled to run as pipelines. The features and models your pipelines produce are managed by a serverless feature store \u002F model registry. We will also show you how to build a UI for your prediction service by writing Python and some HTML.\n\nRead \u003Ca href=\"https:\u002F\u002Fwww.serverless-ml.org\u002Fwhat-is-serverless-machine-learning\">this article\u003C\u002Fa> for an overview on serverless machine learning.\n\n**Prerequisites:** Python - Pandas - Github \n\n# **Modules**\n- ## **Module 00** - Introduction and optional content.\n   - Why Serverless ML: [Video](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=zM2_m898P5g) | [Slides](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F15gwryDoHq88tgxu8CoCbTqr5L9YN9O5p\u002Fview?usp=sharing)\n   - Introduction to the course: [Video](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=FM1YkIl1wXI&list=PLMeDf8qRRqgU_-erq30v-k8_it4pOqhoQ&index=3) | [slides](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F1a5uZHhVSUyxxjrESFea9vONovKROra4L\u002Fview?usp=sharing)\n   - Development Environment & Platforms [Video](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=9kNjky0MQtc&list=PLMeDf8qRRqgU_-erq30v-k8_it4pOqhoQ&index=3) | [slides](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F1LTTHkwV8RirYaz1MeZtoYgTc9TRSrBwr\u002Fview?usp=sharing)\n\n   - ***Introduction to Machine Learning (ML 101)*** [Video](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=RmAGTZ7dy58&list=PLMeDf8qRRqgU_-erq30v-k8_it4pOqhoQ&index=4) | [slides](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F1HXsrSRPcBMW53lgnBnYb95m5eS9oLqRk\u002Fview?usp=sharing)\n   \n- ## **Module 01** - Pandas and ML Pipelines in Python. Write your first serverless App.\n  - Full Lecture: [Video](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=j-XnCflCc0I) | [Slides](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F1L8DHGC5xo0NlNe8xfh4xf4NZV1CEGBA6\u002Fview?usp=sharing)\n\n  - [Lab](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=zAD3miW0Og0) | [Slides](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F1hve9nVrImRhNE8lE26zPcr3X1DDDk7uD\u002Fview?usp=sharing)  | [Homework form](https:\u002F\u002Fforms.gle\u002F2p5odBdpAqvavH1T7)\n  \n- ## **Module 02** - Data modeling and the Feature Store. The Credit-card fraud prediction service. \n    - Full Lecture: [Video](https:\u002F\u002Fyoutu.be\u002FtpxZh8lbcBk) | [Slides](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F1HgAKsHnOms1XCtl_KIEuELudTLtDkhxk\u002Fview?usp=sharing)\n    \n    - [Lab](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=niPayagVxFg) | [Slides](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F1_1oDN5nfpWSUpKNlls45HLllQ75yAWd-\u002Fview?usp=sharing) | [Homework form](https:\u002F\u002Fforms.gle\u002F5g9XtaeBEigKEirGA)\n- ## **Module 03** - Training Pipelines, Inference Pipelines, and the Model Registry.\n    - Full lecture: [Video](https:\u002F\u002Fyoutu.be\u002FBD1UOJs1Bvo) | [Slides](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F1XhfnH7DzwDqQKS6WxDVqWFFas0fi_jnJ\u002Fview?usp=sharing)\n\n    - [Lab](https:\u002F\u002Fyoutu.be\u002FQfzrKgLqEXc) | [Slides](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F1jITx5HGh2uM5vAeknvCaeN6ZPOc2i8AS\u002Fview?usp=sharing)\n- ## **Module 04** - Serverless User Interfaces for Machine Learning Systems.\n    - Full lecture: [Video](https:\u002F\u002Fyoutu.be\u002FGgwIspMUovM) | [Slides](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F10JzJCDwi6IPnJNZ0iApzbwACkAn3C9Y9\u002Fview?usp=sharing)\n\n    - [Lab](https:\u002F\u002Fyoutu.be\u002FsMhCXwm_Wmw) | [Slides](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F1bASaZN68__Ut0RnSuTvhF8LKn240UPtE\u002Fview?usp=sharing)\n\n- ## **Module 05** - Principles and Practices of MLOps\n    - Part 01: [Video](https:\u002F\u002Fyoutu.be\u002F-vbLMtfoBeo) | [Slides](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F1orKJJ2e_1pNgF8X6CFBUKw7qEDQVoVqt\u002Fview?usp=share_link)\n    - Part 02:  [Video](https:\u002F\u002Fyoutu.be\u002Fj4wZmywPs1E) | [Slides](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F13r1OvuuV6Snq1r5PmAvwHU0dTiExQ4iE\u002Fview?usp=share_link)\n    - Lab: [Video](https:\u002F\u002Fyoutu.be\u002FBaAbiFsx25E) | [Slides](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F1WOahxd4s9_NVr8JUUVJvvUFU6ea9konS\u002Fview?usp=share_link)\n\n- ## **Module 06** -Operational machine learning systems: Real-time Machine Learning. \n    - Full lecture: [Video](https:\u002F\u002Fyoutu.be\u002FGEgiIh9a048) | [Slides](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F1VXU2jxEUMIvIY_Xe7XSrNuy0yxXt8glP\u002Fview?usp=share_link)\n    - Lab: [Video](https:\u002F\u002Fyoutu.be\u002FDsyNk3A6ouA) | [Slides](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F1nZJKFMvAFoAu4s5smuc-rIb9EBQVME0Z\u002Fview?usp=share_link)\n\n\n---\n\n## **Learning Outcomes:**\n- Learn to develop and operate AI-enabled (prediction) services on serverless infrastructure\n- Develop and run serverless feature pipelines \n- Deploy features and models to serverless infrastructure\n- Train models and and run batch\u002Finference pipelines\n- Develop a serverless UI for your prediction service\n- Learn MLOps fundamentals: versioning, testing, data validation, and operations\n- Develop and run a real-time serverless machine learning system\n\n## **Course Contents:**\n- Pandas and ML Pipelines in Python. Write your first serverless App.\n- The Feature Store for Machine Learning. Feature engineering for a credit-card fraud serverless App.\n- Training Pipelines and Inference Pipelines\n- Bring a Prediction Service to Life with a User Interface (Gradio, Github Pages, Streamlit)\n- Automated Testing and Versioning of features and models\n- Real-time serverless machine learning systems. Project presentation.\n\n## **Who is the target audience?**\nYou have taken a course in machine learning (ML) and you can program in Python. You want to take the next step beyond training models on static datasets in notebooks. You want to be able to build a prediction service around your model. Maybe you work at an Enterprise and want to demonstrate your models’ value to stakeholders in the stakeholder's own language. Maybe you want to include ML in an existing application or system.\n\n## **Why is this course different?**\nYou don’t need any operations experience beyond using GitHub and writing Python code. You will learn the essentials of MLOps: versioning artifacts, testing artifacts, validating artifacts, and monitoring and upgrading running systems. You will work with raw and live data - you will need to engineer features in pipelines. You will learn how to select, extract, compute, and transform features.\n\n## **Will this course cost me money?**\nNo. You will become a serveless machine learning engineer without having to pay to run your serverless pipelines or to manage your features\u002Fmodels\u002Fuser-interface. We will use Github Actions and Hopsworks that both have generous time-unlimited free tiers.  \n\n**Register now at [Serveless ML Course](https:\u002F\u002Fwww.serverless-ml.org\u002Fregister)** \n\n## **Timeline**\n_Self-paced_\n\n## **Requirements**\n- **Python** environment include a notebook (Jupyter or Colaboratory)\n- https:\u002F\u002Fgithub.com  account\n- https:\u002F\u002Fhopsworks.ai  account \n\n# **Key Technologies**\n\n## **Development environment**\nYou can write, test, debug, and train your models in some Python IDE. We will focus on notebooks and Python programs. You can use Jupyter notebooks or Colaboratory.\n\n## **Github**\nGithub to manage your code, GitHub Actions to run your workflows, and Github Pages for your user interface for non-interactive applications. Github Actions offers a free tier of 500 MB and 2,000 minutes to run your pipelines.\nhttps:\u002F\u002Fdocs.github.com\u002Fen\u002Fbilling\u002Fmanaging-billing-for-github-actions\u002Fabout-billing-for-github-actions\n\n## **Hopsworks**\n[Hopsworks.ai](https:\u002F\u002Fapp.hopsworks.ai) has a free tier of 10 GB of storage.\n\u003Cbr\u002F>\u003Cbr\u002F>\n\n---\n\n## **Useful Resources**\n| name | Description | link |\n|------|-------------|------|\n|**Awesome MLOps**| A collection of links and resources for MLOps| https:\u002F\u002Fgithub.com\u002Fvisenger\u002Fawesome-mlops| \n|**Machine Learning Ops**| a collection of resources on how to facilitate Machine Learning Ops with GitHub.| https:\u002F\u002Fmlops.githubapp.com\u002F|\n|**MLOps Toys**| A curated list of MLOps projects.|https:\u002F\u002Fmlops.toys\u002F|\n|**MLOps Zoomcamp**| teaches practical aspects of productionizing ML services.|https:\u002F\u002Fgithub.com\u002FDataTalksClub\u002Fmlops-zoomcamp| \n|**PYSLACKERS**|A large open community for Python programming enthusiasts.|https:\u002F\u002Fpyslackers.com\u002Fweb| \n|**Feature Store Org**|An open community for everything feature stores.|https:\u002F\u002Fwww.featurestore.org| \n\n\n## **Other MLOps Courses**\n| name | Description | link |\n|------|-------------|------|\n|**MlOps Zoomcamp**| DevOps style course with Python and Docker as prerequisites.| https:\u002F\u002Fgithub.com\u002FDataTalksClub\u002Fmlops-zoomcamp |\n|**Full Stack Deep Learning**| This course shares best practices for the full stack; topics range from problem selection to dataset management to monitoring.| https:\u002F\u002Ffullstackdeeplearning.com\u002F| \n|**MLOps course**| A series of lessons teaching how to apply ML to build production-grade products (by Goku Mohandas).|https:\u002F\u002Fgithub.com\u002FGokuMohandas\u002Fmlops-course |\n\n---\n\n# **Definitions**\n\n- [Context windows for LLMs](http:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fcontext-window-for-llms)\n- [Compound AI Systems](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fcompound-ai-systems)\n- [Feature Store](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Ffeature-store)\n- [Feature Monitoring](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Ffeature-monitoring)\n- [Feature Data](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Ffeature-data) \n- [Flash Attention](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fflash-attention)\n- [Function Calling with LLMs](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Ffunction-calling-with-llms)\n- [Gradient Accumulation](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fgradient-accumulation)\n- [In Context Learning (ICL)](http:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fin-context-learning-icl)\n- [KServe](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fkserve)\n- [ML Logs](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fmachine-learning-logs)\n- [ML Infrastructure](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fmachine-learning-infrastructure)\n- [ML Observability](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fmachine-learning-observability)\n- [ML Pipeline](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fml-pipeline)\n- [ML Systems](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fml-systems)\n- [Model Deployment](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fmodel-deployment)\n- [Model Monitoring](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fmodel-monitoring)\n- [Model Registry](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fmodel-registry)\n- [Model Serving](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fmodel-serving)\n- [PagedAttention](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fpagedattention)\n- [Prompt Store](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fprompt-store)\n- [Retrieval Augmented Generation (RAG) LLM](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fretrieval-augmented-generation-llm)\n- [RoPE Scaling](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Frope-scaling)\n- [Sample Packing](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fsample-packing)\n- [Similarity Search](http:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fsimilarity-search)\n\n# **Support and Partners**\n\u003C\u002Fbr>\n\u003Cp align=\"center\">\n    \u003Ca href=\"https:\u002F\u002Fwww.featurestore.org\">\n        \u003Cimg src=\"https:\u002F\u002Fuploads-ssl.webflow.com\u002F5f32a0dcc815d2a49c58481a\u002F61d5bd3ac0dfb6cabe8d5ebc_FS%20Logo%20.svg\" alt=\"FSorg\" width=\"105\"\u002F>\n    \u003C\u002Fa>\n\u003C\u002Fp>\n\u003Cbr\u002F>\n\u003Cp align=\"center\">\n    \u003Ca href=\"https:\u002F\u002Fwww.hopsworks.ai\">\n        \u003Cimg src=\"https:\u002F\u002Fassets.website-files.com\u002F5f6353590bb01cacbcecfbac\u002F6202a13e7cafec5553703f6b_logo.svg\" alt=\"Hopsworks\" width=\"250\"\u002F>\n    \u003C\u002Fa>\n\u003C\u002Fp>\n","![readme header](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ffeaturestoreorg_serverless-ml-course_readme_c0a46b7768d1.jpg)\n\n\n\u003Ca href=\"https:\u002F\u002Fjoin.slack.com\u002Ft\u002Ffeaturestoreorg\u002Fshared_invite\u002Fzt-ssh8dec1-IsvwLdTsRMssjT4~Ru2RKg\" alt=\"slack\"> \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FJoin Slack-blue.svg?logo=slack\" \u002F>\u003C\u002Fa> \n\u003Ca href=\"https:\u002F\u002Fwww.youtube.com\u002Fchannel\u002FUC-LrK8ei6w57RmKeswkU23Q\" alt=\"youtube\"> \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FYoutube-red.svg?logo=Youtube\" \u002F>\u003C\u002Fa>\n\u003Ca href=\"https:\u002F\u002Fdiscord.com\u002Fchannels\u002F1065622064800735282\u002F1077878719881945109\" alt=\"discord\"> \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FJoin discord-white.svg?logo=discord\" \u002F>\u003C\u002Fa> \n\n\n\n# **[超越笔记本——无服务器机器学习](https:\u002F\u002Fwww.serverless-ml.org)**\n***使用Python构建批处理与实时预测服务***\n\n![无服务器架构](\u002Fassets\u002Fimages\u002Fserverless-ml-architecture.svg \"Serverless Architecture\")\n\n# **概述**\n你无需成为Kubernetes或云计算方面的专家，就能借助机器学习模型构建一个用于做出智能决策的端到端服务。无服务器机器学习（ML）让构建利用机器学习模型进行预测的系统变得简单易行。\n\n在无服务器ML中，你不需要安装、升级或运维任何系统。你只需能够编写可被调度为流水线运行的Python程序即可。你的流水线所生成的特征和模型将由无服务器特征存储库\u002F模型注册表来管理。我们还将向你展示如何通过编写Python代码和一些HTML来为你的预测服务构建用户界面。\n\n阅读 \u003Ca href=\"https:\u002F\u002Fwww.serverless-ml.org\u002Fwhat-is-serverless-machine-learning\">这篇文章\u003C\u002Fa>以了解无服务器机器学习的概览。\n\n**先决条件：** Python - Pandas - Github \n\n# **模块**\n- ## **模块00** - 介绍及选修内容。\n   - 为什么选择无服务器ML：[视频](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=zM2_m898P5g) | [幻灯片](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F15gwryDoHq88tgxu8CoCbTqr5L9YN9O5p\u002Fview?usp=sharing)\n   - 课程简介：[视频](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=FM1YkIl1wXI&list=PLMeDf8qRRqgU_-erq30v-k8_it4pOqhoQ&index=3) | [幻灯片](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F1a5uZHhVSUyxxjrESFea9vONovKROra4L\u002Fview?usp=sharing)\n   - 开发环境与平台 [视频](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=9kNjky0MQtc&list=PLMeDf8qRRqgU_-erq30v-k8_it4pOqhoQ&index=3) | [幻灯片](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F1LTTHkwV8RirYaz1MeZtoYgTc9TRSrBwr\u002Fview?usp=sharing)\n\n   - ***机器学习入门（ML 101）*** [视频](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=RmAGTZ7dy58&list=PLMeDf8qRRqgU_-erq30v-k8_it4pOqhoQ&index=4) | [幻灯片](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F1HXsrSRPcBMW53lgnBnYb95m5eS9oLqRk\u002Fview?usp=sharing)\n   \n- ## **模块01** - Python中的Pandas与ML流水线。编写你的第一个无服务器应用。\n  - 完整讲座：[视频](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=j-XnCflCc0I) | [幻灯片](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F1L8DHGC5xo0NlNe8xfh4xf4NZV1CEGBA6\u002Fview?usp=sharing)\n\n  - [实验](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=zAD3miW0Og0) | [幻灯片](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F1hve9nVrImRhNE8lE26zPcr3X1DDDk7uD\u002Fview?usp=sharing)  | [作业表单](https:\u002F\u002Fforms.gle\u002F2p5odBdpAqvavH1T7)\n  \n- ## **模块02** - 数据建模与特征存储库。信用卡欺诈预测服务。 \n    - 完整讲座：[视频](https:\u002F\u002Fyoutu.be\u002FtpxZh8lbcBk) | [幻灯片](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F1HgAKsHnOms1XCtl_KIEuELudTLtDkhxk\u002Fview?usp=sharing)\n    \n    - [实验](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=niPayagVxFg) | [幻灯片](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F1_1oDN5nfpWSUpKNlls45HLllQ75yAWd-\u002Fview?usp=sharing) | [作业表单](https:\u002F\u002Fforms.gle\u002F5g9XtaeBEigKEirGA)\n- ## **模块03** - 训练流水线、推理流水线与模型注册表。\n    - 完整讲座：[视频](https:\u002F\u002Fyoutu.be\u002FBD1UOJs1Bvo) | [幻灯片](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F1XhfnH7DzwDqQKS6WxDVqWFFas0fi_jnJ\u002Fview?usp=sharing)\n\n    - [实验](https:\u002F\u002Fyoutu.be\u002FQfzrKgLqEXc) | [幻灯片](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F1jITx5HGh2uM5vAeknvCaeN6ZPOc2i8AS\u002Fview?usp=sharing)\n- ## **模块04** - 机器学习系统的无服务器用户界面。\n    - 完整讲座：[视频](https:\u002F\u002Fyoutu.be\u002FGgwIspMUovM) | [幻灯片](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F10JzJCDwi6IPnJNZ0iApzbwACkAn3C9Y9\u002Fview?usp=sharing)\n\n    - [实验](https:\u002F\u002Fyoutu.be\u002FsMhCXwm_Wmw) | [幻灯片](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F1bASaZN68__Ut0RnSuTvhF8LKn240UPtE\u002Fview?usp=sharing)\n\n- ## **模块05** - MLOps的原则与实践\n    - 第一部分：[视频](https:\u002F\u002Fyoutu.be\u002F-vbLMtfoBeo) | [幻灯片](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F1orKJJ2e_1pNgF8X6CFBUKw7qEDQVoVqt\u002Fview?usp=share_link)\n    - 第二部分：[视频](https:\u002F\u002Fyoutu.be\u002Fj4wZmywPs1E) | [幻灯片](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F13r1OvuuV6Snq1r5PmAvwHU0dTiExQ4iE\u002Fview?usp=share_link)\n    - 实验：[视频](https:\u002F\u002Fyoutu.be\u002FBaAbiFsx25E) | [幻灯片](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F1WOahxd4s9_NVr8JUUVJvvUFU6ea9konS\u002Fview?usp=share_link)\n\n- ## **模块06** - 运营级机器学习系统：实时机器学习。 \n    - 完整讲座：[视频](https:\u002F\u002Fyoutu.be\u002FGEgiIh9a048) | [幻灯片](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F1VXU2jxEUMIvIY_Xe7XSrNuy0yxXt8glP\u002Fview?usp=share_link)\n    - 实验：[视频](https:\u002F\u002Fyoutu.be\u002FDsyNk3A6ouA) | [幻灯片](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F1nZJKFMvAFoAu4s5smuc-rIb9EBQVME0Z\u002Fview?usp=share_link)\n\n\n---\n\n## **学习成果：**\n- 学习如何在无服务器基础设施上开发并运营具备AI功能的（预测）服务\n- 开发并运行无服务器特征流水线\n- 将特征和模型部署到无服务器基础设施\n- 训练模型并运行批处理\u002F推理流水线\n- 为你的预测服务开发无服务器用户界面\n- 学习MLOps基础：版本控制、测试、数据验证和运维\n- 开发并运行实时无服务器机器学习系统\n\n## **课程内容：**\n- Python中的Pandas与ML流水线。编写你的第一个无服务器应用。\n- 用于机器学习的特征存储库。为信用卡欺诈无服务器应用进行特征工程。\n- 训练流水线与推理流水线\n- 通过用户界面（Gradio、Github Pages、Streamlit）使预测服务栩栩如生\n- 特征和模型的自动化测试与版本控制\n- 实时无服务器机器学习系统。项目展示。\n\n## **目标受众是谁？**\n你已经学过机器学习（ML）课程，并且会用Python编程。你希望在笔记本中对静态数据集进行模型训练之外更进一步。你想围绕自己的模型构建一个预测服务。也许你在企业工作，想用利益相关者熟悉的方式向他们展示模型的价值。又或者你希望将机器学习集成到现有的应用程序或系统中。\n\n## **这门课程有何不同？**\n你无需具备除使用 GitHub 和编写 Python 代码之外的任何运维经验。你将学习 MLOps 的核心内容：版本化工件、测试工件、验证工件，以及监控和升级运行中的系统。你将处理原始数据和实时数据——你需要在流水线中进行特征工程。你还将学习如何选择、提取、计算和转换特征。\n\n## **这门课程需要付费吗？**\n不需要。你将成为一名无服务器机器学习工程师，而无需为运行你的无服务器流水线或管理你的特征\u002F模型\u002F用户界面支付任何费用。我们将使用 GitHub Actions 和 Hopsworks，它们都提供慷慨的、时间不限的免费层级。\n\n**立即注册 [无服务器机器学习课程](https:\u002F\u002Fwww.serverless-ml.org\u002Fregister)**\n\n## **时间安排**\n_自定进度_\n\n## **要求**\n- 包含笔记本（Jupyter 或 Colaboratory）的 **Python** 环境\n- GitHub 账户\n- Hopsworks.ai 账户\n\n# **关键技术**\n\n## **开发环境**\n你可以在任何 Python IDE 中编写、测试、调试和训练模型。我们将重点放在笔记本和 Python 程序上。你可以使用 Jupyter Notebook 或 Colaboratory。\n\n## **GitHub**\nGitHub 用于管理你的代码，GitHub Actions 用于运行你的工作流，而 GitHub Pages 则适用于非交互式应用的用户界面。GitHub Actions 提供一个免费层级，包含 500 MB 存储空间和 2,000 分钟的流水线运行时间。\nhttps:\u002F\u002Fdocs.github.com\u002Fen\u002Fbilling\u002Fmanaging-billing-for-github-actions\u002Fabout-billing-for-github-actions\n\n## **Hopsworks**\n[Hopsworks.ai](https:\u002F\u002Fapp.hopsworks.ai) 提供 10 GB 存储空间的免费层级。\n\u003Cbr\u002F>\u003Cbr\u002F>\n\n---\n\n## **实用资源**\n| 名称 | 描述 | 链接 |\n|------|-------------|------|\n|**Awesome MLOps**| MLOps 相关链接和资源的集合| https:\u002F\u002Fgithub.com\u002Fvisenger\u002Fawesome-mlops| \n|**Machine Learning Ops**| 关于如何利用 GitHub 推动机器学习运维的资源集合。| https:\u002F\u002Fmlops.githubapp.com\u002F|\n|**MLOps Toys**| 精选的 MLOps 项目列表。|https:\u002F\u002Fmlops.toys\u002F|\n|**MLOps Zoomcamp**| 讲授将机器学习服务投入生产环境的实用技巧。|https:\u002F\u002Fgithub.com\u002FDataTalksClub\u002Fmlops-zoomcamp| \n|**PYSLACKERS**| 一个面向 Python 编程爱好者的大型开放社区。|https:\u002F\u002Fpyslackers.com\u002Fweb| \n|**Feature Store Org**| 一个关于特征存储的开放社区。|https:\u002F\u002Fwww.featurestore.org| \n\n\n## **其他 MLOps 课程**\n| 名称 | 描述 | 链接 |\n|------|-------------|------|\n|**MlOps Zoomcamp**| 一门以 DevOps 为导向的课程，前提条件是掌握 Python 和 Docker。| https:\u002F\u002Fgithub.com\u002FDataTalksClub\u002Fmlops-zoomcamp |\n|**Full Stack Deep Learning**| 本课程分享全栈最佳实践；主题涵盖从问题选择到数据集管理再到监控等各个方面。| https:\u002F\u002Ffullstackdeeplearning.com\u002F| \n|**MLOps 课程**| 由 Goku Mohandas 主讲的一系列课程，教授如何应用机器学习构建生产级产品。|https:\u002F\u002Fgithub.com\u002FGokuMohandas\u002Fmlops-course |\n\n---\n\n# **定义**\n\n- [LLM 的上下文窗口](http:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fcontext-window-for-llms)\n- [复合 AI 系统](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fcompound-ai-systems)\n- [特征存储](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Ffeature-store)\n- [特征监控](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Ffeature-monitoring)\n- [特征数据](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Ffeature-data) \n- [Flash Attention](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fflash-attention)\n- [LLM 的函数调用](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Ffunction-calling-with-llms)\n- [梯度累积](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fgradient-accumulation)\n- [上下文学习 (ICL)](http:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fin-context-learning-icl)\n- [KServe](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fkserve)\n- [机器学习日志](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fmachine-learning-logs)\n- [机器学习基础设施](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fmachine-learning-infrastructure)\n- [机器学习可观测性](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fmachine-learning-observability)\n- [机器学习流水线](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fml-pipeline)\n- [机器学习系统](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fml-systems)\n- [模型部署](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fmodel-deployment)\n- [模型监控](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fmodel-monitoring)\n- [模型注册表](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fmodel-registry)\n- [模型推理服务](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fmodel-serving)\n- [分页注意力](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fpagedattention)\n- [提示存储](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fprompt-store)\n- [检索增强生成 (RAG) LLM](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fretrieval-augmented-generation-llm)\n- [RoPE 缩放](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Frope-scaling)\n- [样本打包](https:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fsample-packing)\n- [相似性搜索](http:\u002F\u002Fwww.hopsworks.ai\u002Fdictionary\u002Fsimilarity-search)\n\n# **支持与合作伙伴**\n\u003C\u002Fbr>\n\u003Cp align=\"center\">\n    \u003Ca href=\"https:\u002F\u002Fwww.featurestore.org\">\n        \u003Cimg src=\"https:\u002F\u002Fuploads-ssl.webflow.com\u002F5f32a0dcc815d2a49c58481a\u002F61d5bd3ac0dfb6cabe8d5ebc_FS%20Logo%20.svg\" alt=\"FSorg\" width=\"105\"\u002F>\n    \u003C\u002Fa>\n\u003C\u002Fp>\n\u003Cbr\u002F>\n\u003Cp align=\"center\">\n    \u003Ca href=\"https:\u002F\u002Fwww.hopsworks.ai\">\n        \u003Cimg src=\"https:\u002F\u002Fassets.website-files.com\u002F5f6353590bb01cacbcecfbac\u002F6202a13e7cafec5553703f6b_logo.svg\" alt=\"Hopsworks\" width=\"250\"\u002F>\n    \u003C\u002Fa>\n\u003C\u002Fp>","# Serverless ML Course 快速上手指南\n\n本指南旨在帮助开发者无需掌握 Kubernetes 或复杂的云运维知识，仅通过编写 Python 代码即可构建端到端的批量与实时机器学习预测服务。\n\n## 环境准备\n\n在开始之前，请确保你具备以下基础环境和账号：\n\n### 1. 前置技能\n- **Python**: 熟悉 Python 编程。\n- **Pandas**: 了解基本的数据处理操作。\n- **Git**: 熟悉基本的 GitHub 操作。\n\n### 2. 系统要求\n- **开发环境**: 本地安装的 Python 环境，或使用在线 Notebook 环境（推荐 **Jupyter Notebook** 或 **Google Colaboratory**）。\n- **网络连接**: 能够访问 GitHub 和 Hopsworks 平台。\n\n### 3. 必需账号\n本课程完全免费，利用各平台的免费层级即可运行，无需信用卡。请提前注册以下账号：\n- **GitHub**: [https:\u002F\u002Fgithub.com](https:\u002F\u002Fgithub.com)\n  - 用于代码管理、使用 GitHub Actions 运行流水线以及部署 UI (GitHub Pages)。\n- **Hopsworks**: [https:\u002F\u002Fapp.hopsworks.ai](https:\u002F\u002Fapp.hopsworks.ai)\n  - 作为无服务器特征存储（Feature Store）和模型注册表（Model Registry）。注册后可获得 10GB 免费存储空间。\n\n---\n\n## 安装步骤\n\n本课程基于云端无服务器架构，**无需在本地安装复杂的服务器软件或 Docker**。主要配置步骤如下：\n\n### 1. 克隆课程仓库\n打开终端，克隆项目代码到本地：\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002Flogicalclocks\u002Fserverless-ml-course.git\ncd serverless-ml-course\n```\n\n### 2. 配置 Python 依赖\n进入对应的模块目录（例如 Module 01），安装所需的 Python 库：\n\n```bash\npip install -r requirements.txt\n```\n*注：如果在 Google Colab 中运行，请使用 `!pip install -r requirements.txt`。*\n\n### 3. 配置 Hopsworks API Key\n1. 登录 [Hopsworks AI](https:\u002F\u002Fapp.hopsworks.ai)。\n2. 在用户设置中找到 **API Keys** 并生成一个新的 Key。\n3. 在你的 Python 脚本或 Notebook 中配置该 Key（通常在代码初始化部分）：\n\n```python\nimport hopsworks\n\n# 将 'YOUR_API_KEY' 替换为你生成的实际 Key\nproject = hopsworks.login(api_key_value=\"YOUR_API_KEY\")\n```\n\n### 4. 配置 GitHub Actions (可选，用于自动化流水线)\n若需自动运行训练或推理流水线：\n1. 在 GitHub 仓库的 **Settings** > **Secrets and variables** > **Actions** 中添加新的 Secret。\n2. Name: `HOPSWORKS_API_KEY`\n3. Value: 填入你的 Hopsworks API Key。\n\n---\n\n## 基本使用\n\n以下是一个最简单的示例，展示如何连接到特征存储并读取数据（对应课程 Module 01 & 02 的核心逻辑）。\n\n### 示例：连接特征存储并获取数据\n\n创建一个名为 `first_app.py` 的文件或在 Notebook 中运行以下代码：\n\n```python\nimport hopsworks\nimport pandas as pd\n\n# 1. 登录 Hopsworks 项目\n# 确保已设置环境变量 HOPSWORKS_API_KEY 或在 login() 中传入 api_key_value\nproject = hopsworks.login()\n\n# 2. 获取特征存储实例\nfs = project.get_feature_store()\n\n# 3. 获取已有的特征组 (Feature Group) \n# 这里以课程中的信用卡欺诈数据集为例\nfeature_group = fs.get_feature_group(name=\"credit_card_transactions\", version=1)\n\n# 4. 读取数据为 Pandas DataFrame\ndf = feature_group.read()\n\n# 5. 查看数据前几行\nprint(df.head())\n\n# 6. (可选) 简单的数据预处理示例\ndf['amount_normalized'] = df['amount'] \u002F df['amount'].max()\n\nprint(\"数据预处理完成，准备进入下一步流水线...\")\n```\n\n### 运行你的第一个流水线\n完成代码编写后，你可以：\n1. **本地运行**: 直接在终端执行 `python first_app.py`。\n2. **云端调度**: 将代码推送到 GitHub，配置 `.github\u002Fworkflows` 文件，利用 **GitHub Actions** 自动触发该 Python 脚本作为无服务器流水线运行。\n\n### 下一步\n按照课程模块顺序深入学习：\n- **Module 01**: 构建第一个 Serverless App。\n- **Module 02**: 学习特征工程与特征存储。\n- **Module 03**: 创建训练与推理流水线。\n- **Module 04**: 使用 Gradio 或 Streamlit 构建预测服务 UI。","某金融科技公司数据团队急需将信用卡欺诈检测模型从实验阶段推向生产环境，以提供实时风险拦截服务。\n\n### 没有 serverless-ml-course 时\n- **运维门槛极高**：团队成员必须精通 Kubernetes 和云基础设施架构，否则无法搭建高可用的预测服务，导致项目长期卡在部署环节。\n- **资源管理混乱**：缺乏统一的特征存储和模型注册中心，训练数据与线上推理数据不一致，常出现“模型在笔记本上有效，上线就失效”的问题。\n- **开发流程割裂**：数据科学家需手动协调批处理与实时推理管道，编写大量样板代码来调度任务，难以专注于核心算法优化。\n- **前端集成困难**：为预测服务构建监控或交互界面需要额外招募全栈工程师，沟通成本高且迭代缓慢。\n\n### 使用 serverless-ml-course 后\n- **零运维负担**：团队仅需编写 Python 脚本即可自动编排批处理和实时管道，无需安装、升级或操作任何底层服务器系统。\n- **数据一致性保障**：通过课程指导建立的无服务器特征存储和模型注册表，自动管理特征与模型版本，确保训练与推理环境高度一致。\n- **端到端流水线化**：利用标准化的训练与推理管道模板，快速将 Pandas 数据处理逻辑转化为可调度的生产级应用，大幅缩短上线周期。\n- **全栈能力下沉**：数据工程师可直接用 Python 配合少量 HTML 构建预测服务的前端 UI，独立实现从模型到用户界面的完整闭环。\n\nserverless-ml-course 让不懂复杂云架构的团队也能仅凭 Python 技能，低成本构建出具备批量与实时预测能力的企业级 AI 服务。","https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ffeaturestoreorg_serverless-ml-course_c0a46b77.jpg","featurestoreorg","feature stores for ml","https:\u002F\u002Foss.gittoolsai.com\u002Favatars\u002Ffeaturestoreorg_0811a1a7.png","The github repository for all things part of the feature store community. ",null,"featurestoreorg@gmail.com","featurestore.org","https:\u002F\u002Fgithub.com\u002Ffeaturestoreorg",[84,88,92],{"name":85,"color":86,"percentage":87},"Jupyter Notebook","#DA5B0B",90.7,{"name":89,"color":90,"percentage":91},"Python","#3572A5",9.3,{"name":93,"color":94,"percentage":95},"Shell","#89e051",0.1,684,301,"2026-04-02T15:36:58","CC0-1.0","未说明",{"notes":102,"python":103,"dependencies":104},"本课程主要基于云端无服务器架构，无需本地安装复杂系统。核心要求是拥有 GitHub 账号（用于代码管理和 GitHub Actions 运行流水线）和 Hopsworks.ai 账号（用于特征存储和模型注册，提供免费层级）。开发环境可使用本地 Python IDE、Jupyter Notebook 或 Google Colaboratory。课程强调无需 Kubernetes 或云计算专家知识，仅需编写 Python 代码即可构建批处理和实时预测服务。","未说明 (需具备 Python 编程能力)",[105,106,107,108,109],"pandas","github-actions","hopsworks","gradio","streamlit",[13],[112,113,114,115,116,117,118,119],"mlops","serverless","course","feature-engineering","feature-store","machine-learning","ml","model-deployment","2026-03-27T02:49:30.150509","2026-04-06T09:44:37.102035",[],[]]