[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"similar-Ramakm--ai-hands-on":3,"tool-Ramakm--ai-hands-on":64},[4,17,27,35,44,52],{"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":10,"last_commit_at":41,"category_tags":42,"status":16},4292,"Deep-Live-Cam","hacksider\u002FDeep-Live-Cam","Deep-Live-Cam 是一款专注于实时换脸与视频生成的开源工具，用户仅需一张静态照片，即可通过“一键操作”实现摄像头画面的即时变脸或制作深度伪造视频。它有效解决了传统换脸技术流程繁琐、对硬件配置要求极高以及难以实时预览的痛点，让高质量的数字内容创作变得触手可及。\n\n这款工具不仅适合开发者和技术研究人员探索算法边界，更因其极简的操作逻辑（仅需三步：选脸、选摄像头、启动），广泛适用于普通用户、内容创作者、设计师及直播主播。无论是为了动画角色定制、服装展示模特替换，还是制作趣味短视频和直播互动，Deep-Live-Cam 都能提供流畅的支持。\n\n其核心技术亮点在于强大的实时处理能力，支持口型遮罩（Mouth Mask）以保留使用者原始的嘴部动作，确保表情自然精准；同时具备“人脸映射”功能，可同时对画面中的多个主体应用不同面孔。此外，项目内置了严格的内容安全过滤机制，自动拦截涉及裸露、暴力等不当素材，并倡导用户在获得授权及明确标注的前提下合规使用，体现了技术发展与伦理责任的平衡。",88924,"2026-04-06T03:28:53",[13,14,15,43],"视频",{"id":45,"name":46,"github_repo":47,"description_zh":48,"stars":49,"difficulty_score":23,"last_commit_at":50,"category_tags":51,"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":53,"name":54,"github_repo":55,"description_zh":56,"stars":57,"difficulty_score":23,"last_commit_at":58,"category_tags":59,"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",[14,60,43,61,15,62,26,13,63],"数据工具","插件","其他","音频",{"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":80,"owner_email":81,"owner_twitter":82,"owner_website":83,"owner_url":84,"languages":85,"stars":94,"forks":95,"last_commit_at":96,"license":97,"difficulty_score":23,"env_os":98,"env_gpu":98,"env_ram":98,"env_deps":99,"category_tags":111,"github_topics":112,"view_count":23,"oss_zip_url":126,"oss_zip_packed_at":126,"status":16,"created_at":127,"updated_at":128,"faqs":129,"releases":130},4197,"Ramakm\u002Fai-hands-on","ai-hands-on","A group of notebooks  and other files which can help you learn AI from scratch.","ai-hands-on 是一套专为零基础学习者打造的 AI 工程实战指南，通过一系列结构清晰的 Jupyter Notebook，帮助用户从第一性原理出发，系统掌握人工智能核心技术。它解决了传统教程中理论脱离实践、代码碎片化以及缺乏完整项目链路的问题，让学习者能够亲手构建神经网络并理解现代大语言模型（LLM）系统的端到端运作。\n\n这套资源非常适合希望转行 AI 的开发者、需要夯实基础的工程师，以及计算机相关专业的学生。其独特的技术亮点在于“从零构建”的教学理念：不仅涵盖数学基础与 PyTorch 操作，更引导用户不使用高级封装库，而是从头编写神经元、层、优化器乃至完整的 Transformer 架构。此外，项目还深入讲解了检索增强生成（RAG）流水线搭建与光学字符识别（OCR）等前沿应用场景，并提供了从监督学习到强化学习的经典机器学习模型实现。配合推荐的经典书单与明确的学习路径，ai-hands-on 旨在为用户提供构建真实 AI 系统所需的直觉、结构与清晰度，是通往 AI 工程师之路的优质实践手册。","# AI Engineering: Hands-on\n![Stars](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FRamakm\u002Fai-hands-on?style=flat-square)\n![Forks](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fforks\u002FRamakm\u002Fai-hands-on?style=flat-square)\n![PRs](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fissues-pr\u002FRamakm\u002Fai-hands-on?style=flat-square)\n![Issues](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fissues\u002FRamakm\u002Fai-hands-on?style=flat-square)\n![Contributors](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fcontributors\u002FRamakm\u002Fai-hands-on?style=flat-square)\n![License](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Flicense\u002FRamakm\u002Fai-hands-on?style=flat-square)\n\u003Cimg width=\"2000\" height=\"600\" alt=\"image\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FRamakm_ai-hands-on_readme_c37f227f0382.png\" \u002F>\n\nA complete, hands-on guide to becoming an AI Engineer.\n\nThis repository is designed to help you learn AI from first principles, build real neural networks, and understand modern LLM systems end-to-end.\nYou'll progress through math, PyTorch, deep learning, transformers, RAG, and OCR — with clean, intuitive Jupyter notebooks guiding you at every step.\n\nWhether you're a beginner or an engineer levelling up, this repo gives you the clarity, structure, and intuition needed to build real AI systems.\n\n#### ⭐ Star This Repo\n\nIf you learn something useful, a star is appreciated.\n\n## Repository Structure\n\n### 1. Math Fundamentals\n- Math functions, derivatives, vectors, and gradients\n- Matrix operations and linear algebra\n- Probability and statistics\n\n### 2. PyTorch Basics\n- Creating and manipulating tensors\n- Matrix multiplication, transposing, and reshaping\n- Indexing, slicing, and concatenating tensors\n- Special tensor creation functions\n\n### 3. Neural-Network(NN)\n- Building neurons, layers, and networks from scratch\n- Normalization techniques (RMSNorm)\n- Activation functions\n- Optimizers (Adam, Muon) and learning rate decay\n\n### 4. Transformers\n- Attention and self-attention mechanisms\n- Multi-head attention\n- Decoder-only transformer architecture\n\n### 5. Retrieval-Augmented Generation (RAG)\n- Building RAG pipelines end to end\n- Indexing, retrieval, chunking strategies\n- Integrations with embedding models and vector stores\n\n### 6. Optical Character Recognition (OCR)\n- OCR pipeline and utilities\n- Preprocessing images and extracting text\n\n## Books\n\nRecommended reading to deepen your understanding (not included):\n\n- `AI Engineering` by Chip Huyen\n- `Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow` by Aurélien Géron\n- `Deep Learning` by Ian Goodfellow, Yoshua Bengio, and Aaron Courville\n- `The Elements of Statistical Learning` by Trevor Hastie, Robert Tibshirani, and Jerome Friedman\n- `Neural Networks and Deep Learning` by Michael Nielsen\n- `SQL Cookbook` by Anthony Molinaro\n\nFor more books in AI\u002FML, I have created another repo for this [Check Here](https:\u002F\u002Fgithub.com\u002FRamakm\u002FAI-ML-Book-References.git). I will be adding lot more in coming days\u002Fmonths. If you are interested to read book, go check this repo out.\n\n## Learning Path\n\nFor a recommended step-by-step progression through the materials, see the Learning Path:\n\n- `Start_here\u002Flearning_path.md`\n\n## Requirements\n\nInstall dependencies with:\n```bash\npip install -r requirements.txt\n```\n\nSome subfolders (for example `5.RAG\u002F` and `6.OCR\u002F`) include their own `requirements.txt` with additional dependencies.\n\n## Usage\n\nRecommended workflow:\n\n1. Open Jupyter in the project root:\n   ```bash\n   jupyter lab\n   # or\n   jupyter notebook\n   ```\n2. Work through notebooks in order:\n   - `1.Math\u002F`\n   - `2.PyTorch\u002F`\n   - `3.Neural-Network(NN)\u002F`\n   - `4.Transformer\u002F`\n\n3. Folder to run separately:\n   - `5.RAG\u002F`\n   - `6.OCR\u002F`\n  \n4. Resources\n5. Basic ML Model Implementation (Supervised + Un-supervised + RL)\n   - `1.Linear Regression`\n   - `2.Logistic Regression`\n   - `3.Decision Tree Model`\n   - `4.Naive Bayes Classification`\n  \n## Machine Learning Frameworks\n\n| Tool         | Category          | Link                                                                                     |\n| ------------ | ----------------- | ---------------------------------------------------------------------------------------- |\n| Scikit-learn | Traditional ML    | [https:\u002F\u002Fscikit-learn.org\u002Fstable\u002F](https:\u002F\u002Fscikit-learn.org\u002Fstable\u002F)                     |\n| XGBoost      | Gradient Boosting | [https:\u002F\u002Fxgboost.ai\u002F](https:\u002F\u002Fxgboost.ai\u002F)                                               |\n| LightGBM     | Gradient Boosting | [https:\u002F\u002Flightgbm.readthedocs.io\u002Fen\u002Fstable\u002F](https:\u002F\u002Flightgbm.readthedocs.io\u002Fen\u002Fstable\u002F) |\n| CatBoost     | Gradient Boosting | [https:\u002F\u002Fcatboost.ai\u002F](https:\u002F\u002Fcatboost.ai\u002F)                                             |\n\n\n## GEN AI References\n\n| Resource                              | Focus Area             | Link                                                                                                                                                                   |\n| ------------------------------------- | ---------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------- |\n| Microsoft Generative AI for Beginners | Intro to GenAI         | [https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fgenerative-ai-for-beginners](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fgenerative-ai-for-beginners)                                                   |\n| Generative AI for Everyone            | Non-technical overview | [https:\u002F\u002Fwww.coursera.org\u002Flearn\u002Fgenerative-ai-for-everyone](https:\u002F\u002Fwww.coursera.org\u002Flearn\u002Fgenerative-ai-for-everyone)                                                 |\n| Building Blocks of Generative AI      | Conceptual foundations | [https:\u002F\u002Fshriftman.substack.com\u002Fp\u002Fthe-building-blocks-of-generative](https:\u002F\u002Fshriftman.substack.com\u002Fp\u002Fthe-building-blocks-of-generative)                               |\n| The Illustrated Transformer           | Transformers           | [https:\u002F\u002Fjalammar.github.io\u002Fillustrated-transformer\u002F](https:\u002F\u002Fjalammar.github.io\u002Fillustrated-transformer\u002F)                                                             |\n| LLMs Explained Briefly                | LLM basics video       | [https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=LPZh9BOjkQs](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=LPZh9BOjkQs)                                                                             |\n| Intro to LLMs                         | LLM overview video     | [https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=zjkBMFhNj_g](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=zjkBMFhNj_g)                                                                             |\n| Understanding LLMs                    | Deep dive              | [https:\u002F\u002Fmagazine.sebastianraschka.com\u002Fp\u002Funderstanding-large-language-models](https:\u002F\u002Fmagazine.sebastianraschka.com\u002Fp\u002Funderstanding-large-language-models)             |\n| Visual Guide to Reasoning LLMs        | Reasoning models       | [https:\u002F\u002Fnewsletter.maartengrootendorst.com\u002Fp\u002Fa-visual-guide-to-reasoning-llms](https:\u002F\u002Fnewsletter.maartengrootendorst.com\u002Fp\u002Fa-visual-guide-to-reasoning-llms)         |\n| Understanding Reasoning LLMs          | Reasoning theory       | [https:\u002F\u002Fmagazine.sebastianraschka.com\u002Fp\u002Funderstanding-reasoning-llms](https:\u002F\u002Fmagazine.sebastianraschka.com\u002Fp\u002Funderstanding-reasoning-llms)                           |\n| Understanding Multimodal LLMs         | Vision + text models   | [https:\u002F\u002Fmagazine.sebastianraschka.com\u002Fp\u002Funderstanding-multimodal-llms](https:\u002F\u002Fmagazine.sebastianraschka.com\u002Fp\u002Funderstanding-multimodal-llms)                         |\n| Visual Guide to MoE                   | Mixture of Experts     | [https:\u002F\u002Fnewsletter.maartengrootendorst.com\u002Fp\u002Fa-visual-guide-to-mixture-of-experts](https:\u002F\u002Fnewsletter.maartengrootendorst.com\u002Fp\u002Fa-visual-guide-to-mixture-of-experts) |\n| Finetuning LLMs                       | Model training         | [https:\u002F\u002Fmagazine.sebastianraschka.com\u002Fp\u002Ffinetuning-large-language-models](https:\u002F\u002Fmagazine.sebastianraschka.com\u002Fp\u002Ffinetuning-large-language-models)                   |\n| How Transformer LLMs Work             | Architecture           | [https:\u002F\u002Fwww.deeplearning.ai\u002Fshort-courses\u002Fhow-transformer-llms-work\u002F](https:\u002F\u002Fwww.deeplearning.ai\u002Fshort-courses\u002Fhow-transformer-llms-work\u002F)                           |\n| Build GPT from Scratch                | Hands-on               | [https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=kCc8FmEb1nY](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=kCc8FmEb1nY)                                                                             |\n| LLM Course (GitHub)                   | Structured learning    | [https:\u002F\u002Fgithub.com\u002Fmlabonne\u002Fllm-course](https:\u002F\u002Fgithub.com\u002Fmlabonne\u002Fllm-course)                                                                                       |\n| LLM Course (Hugging Face)             | Practical LLMs         | [https:\u002F\u002Fhuggingface.co\u002Flearn\u002Fllm-course\u002Fchapter1\u002F1](https:\u002F\u002Fhuggingface.co\u002Flearn\u002Fllm-course\u002Fchapter1\u002F1)                                                               |\n| Awesome LLM Apps                      | Project ideas          | [https:\u002F\u002Fgithub.com\u002FShubhamsaboo\u002Fawesome-llm-apps](https:\u002F\u002Fgithub.com\u002FShubhamsaboo\u002Fawesome-llm-apps)                                                                   |\n| How RAG Enhances LLMs                 | RAG                    | [https:\u002F\u002Fawesomeneuron.substack.com\u002Fp\u002Fhow-rag-enhances-llms-a-step-by-step](https:\u002F\u002Fawesomeneuron.substack.com\u002Fp\u002Fhow-rag-enhances-llms-a-step-by-step)                                                       |\n| Visual Guide to AI Agents             | AI Agents              | [https:\u002F\u002Fawesomeneuron.substack.com\u002Fp\u002Fa-visual-guide-to-ai-agents](https:\u002F\u002Fawesomeneuron.substack.com\u002Fp\u002Fa-visual-guide-to-ai-agents)                                        |\n\n## Contributing\n\nContributions are welcome!\n\nPlease ensure:\n\n- Notebooks are clean (Restart & Run All before committing)\n- Existing structure & naming conventions are followed\n- PRs are focused, readable, and documented\n- In folders like RAG and OCR, please maintain the cleaned structure part\n- If you want to add something new folders, make it proper structure way.\n\n## License\n\n- This project is licensed under the MIT License. See `LICENSE` for details.\n\n## Connect with me\n\n[![X](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FX-000000?style=for-the-badge&logo=x&logoColor=white)](https:\u002F\u002Fx.com\u002Ftechwith_ram)\n[![Instagram](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FInstagram-E4405F?style=for-the-badge&logo=instagram&logoColor=white)](https:\u002F\u002Finstagram.com\u002Ftechwith.ram)\n[![GitHub](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FGitHub-181717?style=for-the-badge&logo=github&logoColor=white)](https:\u002F\u002Fgithub.com\u002FRamakm)\n\n","# AI 工程：动手实践\n![Stars](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FRamakm\u002Fai-hands-on?style=flat-square)\n![Forks](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fforks\u002FRamakm\u002Fai-hands-on?style=flat-square)\n![PRs](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fissues-pr\u002FRamakm\u002Fai-hands-on?style=flat-square)\n![Issues](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fissues\u002FRamakm\u002Fai-hands-on?style=flat-square)\n![Contributors](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fcontributors\u002FRamakm\u002Fai-hands-on?style=flat-square)\n![License](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Flicense\u002FRamakm\u002Fai-hands-on?style=flat-square)\n\u003Cimg width=\"2000\" height=\"600\" alt=\"image\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FRamakm_ai-hands-on_readme_c37f227f0382.png\" \u002F>\n\n一本全面的、实践导向的指南，助你成为一名 AI 工程师。\n\n本仓库旨在帮助你从基础原理出发学习 AI，构建真正的神经网络，并端到端地理解现代 LLM 系统。你将逐步深入数学、PyTorch、深度学习、Transformer、RAG 和 OCR 等领域——每一步都有清晰直观的 Jupyter 笔记本引导你前进。\n\n无论你是初学者，还是希望进一步提升技能的工程师，这个仓库都能为你提供构建真实 AI 系统所需的清晰思路、系统化结构和直观理解。\n\n#### ⭐ 给本仓库点个赞\n\n如果你学到了有用的知识，给个 star 表示感谢吧！\n\n## 仓库结构\n\n### 1. 数学基础\n- 数学函数、导数、向量与梯度\n- 矩阵运算与线性代数\n- 概率与统计\n\n### 2. PyTorch 基础\n- 创建和操作张量\n- 矩阵乘法、转置与重塑\n- 张量的索引、切片与拼接\n- 特殊张量创建函数\n\n### 3. 神经网络 (NN)\n- 从零开始构建神经元、层和网络\n- 归一化技术（RMSNorm）\n- 激活函数\n- 优化器（Adam、Muon）及学习率衰减\n\n### 4. Transformer\n- 注意力机制与自注意力机制\n- 多头注意力\n- 解码器-only 架构的 Transformer\n\n### 5. 检索增强生成 (RAG)\n- 端到端构建 RAG 流水线\n- 索引、检索与分块策略\n- 与嵌入模型和向量数据库的集成\n\n### 6. 光学字符识别 (OCR)\n- OCR 流水线与实用工具\n- 图像预处理与文本提取\n\n## 书籍推荐\n\n以下是一些有助于加深理解的推荐读物（未包含在本仓库中）：\n\n- Chip Huyen 的《AI 工程》\n- Aurélien Géron 的《使用 Scikit-Learn、Keras 和 TensorFlow 的机器学习实战》\n- Ian Goodfellow、Yoshua Bengio 和 Aaron Courville 的《深度学习》\n- Trevor Hastie、Robert Tibshirani 和 Jerome Friedman 的《统计学习要素》\n- Michael Nielsen 的《神经网络与深度学习》\n- Anthony Molinaro 的《SQL 烹饪书》\n\n如需更多 AI\u002FML 方面的书籍，我已创建了另一个仓库 [点击这里](https:\u002F\u002Fgithub.com\u002FRamakm\u002FAI-ML-Book-References.git)。未来几天或几个月内，我会继续添加更多内容。如果你对阅读相关书籍感兴趣，不妨去看看这个仓库。\n\n## 学习路径\n\n关于材料的推荐循序渐进学习路径，请参阅：\n\n- `Start_here\u002Flearning_path.md`\n\n## 依赖安装\n\n通过以下命令安装依赖：\n```bash\npip install -r requirements.txt\n```\n\n部分子文件夹（例如 `5.RAG\u002F` 和 `6.OCR\u002F`）包含各自的 `requirements.txt` 文件，用于安装额外的依赖项。\n\n## 使用说明\n\n推荐的工作流程如下：\n\n1. 在项目根目录下打开 Jupyter：\n   ```bash\n   jupyter lab\n   # 或\n   jupyter notebook\n   ```\n2. 按顺序依次学习各笔记本：\n   - `1.Math\u002F`\n   - `2.PyTorch\u002F`\n   - `3.Neural-Network(NN)\u002F`\n   - `4.Transformer\u002F`\n\n3. 需要单独运行的文件夹：\n   - `5.RAG\u002F`\n   - `6.OCR\u002F`\n\n4. 资源部分\n5. 基础机器学习模型实现（监督学习 + 非监督学习 + 强化学习）：\n   - `1.线性回归`\n   - `2.逻辑回归`\n   - `3.决策树模型`\n   - `4.朴素贝叶斯分类`\n\n## 机器学习框架\n\n| 工具         | 类别          | 链接                                                                                     |\n| ------------ | ----------------- | ---------------------------------------------------------------------------------------- |\n| Scikit-learn | 传统机器学习    | [https:\u002F\u002Fscikit-learn.org\u002Fstable\u002F](https:\u002F\u002Fscikit-learn.org\u002Fstable\u002F)                     |\n| XGBoost      | 梯度提升        | [https:\u002F\u002Fxgboost.ai\u002F](https:\u002F\u002Fxgboost.ai\u002F)                                               |\n| LightGBM     | 梯度提升        | [https:\u002F\u002Flightgbm.readthedocs.io\u002Fen\u002Fstable\u002F](https:\u002F\u002Flightgbm.readthedocs.io\u002Fen\u002Fstable\u002F) |\n| CatBoost     | 梯度提升        | [https:\u002F\u002Fcatboost.ai\u002F](https:\u002F\u002Fcatboost.ai\u002F)                                             |\n\n## 生成式AI参考资料\n\n| 资源                              | 重点领域             | 链接                                                                                                                                                                   |\n| ------------------------------------- | ---------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------- |\n| 微软面向初学者的生成式AI | 生成式AI入门         | [https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fgenerative-ai-for-beginners](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fgenerative-ai-for-beginners)                                                   |\n| 每个人都能懂的生成式AI            | 非技术性概述 | [https:\u002F\u002Fwww.coursera.org\u002Flearn\u002Fgenerative-ai-for-everyone](https:\u002F\u002Fwww.coursera.org\u002Flearn\u002Fgenerative-ai-for-everyone)                                                 |\n| 生成式AI的构建模块      | 概念基础 | [https:\u002F\u002Fshriftman.substack.com\u002Fp\u002Fthe-building-blocks-of-generative](https:\u002F\u002Fshriftman.substack.com\u002Fp\u002Fthe-building-blocks-of-generative)                               |\n| 图解Transformer           | Transformer模型           | [https:\u002F\u002Fjalammar.github.io\u002Fillustrated-transformer\u002F](https:\u002F\u002Fjalammar.github.io\u002Fillustrated-transformer\u002F)                                                             |\n| LLM简要解析                | LLM基础知识视频       | [https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=LPZh9BOjkQs](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=LPZh9BOjkQs)                                                                             |\n| LLM入门                   | LLM概述视频     | [https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=zjkBMFhNj_g](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=zjkBMFhNj_g)                                                                             |\n| 理解LLM                   | 深度解析              | [https:\u002F\u002Fmagazine.sebastianraschka.com\u002Fp\u002Funderstanding-large-language-models](https:\u002F\u002Fmagazine.sebastianraschka.com\u002Fp\u002Funderstanding-large-language-models)             |\n| 理性LLM视觉指南        | 理性模型       | [https:\u002F\u002Fnewsletter.maartengrootendorst.com\u002Fp\u002Fa-visual-guide-to-reasoning-llms](https:\u002F\u002Fnewsletter.maartengrootendorst.com\u002Fp\u002Fa-visual-guide-to-reasoning-llms)         |\n| 理解理性LLM          | 理论分析       | [https:\u002F\u002Fmagazine.sebastianraschka.com\u002Fp\u002Funderstanding-reasoning-llms](https:\u002F\u002Fmagazine.sebastianraschka.com\u002Fp\u002Funderstanding-reasoning-llms)                           |\n| 多模态LLM的理解         | 视觉+文本模型   | [https:\u002F\u002Fmagazine.sebastianraschka.com\u002Fp\u002Funderstanding-multimodal-llms](https:\u002F\u002Fmagazine.sebastianraschka.com\u002Fp\u002Funderstanding-multimodal-llms)                         |\n| MoE视觉指南             | 混合专家模型     | [https:\u002F\u002Fnewsletter.maartengrootendorst.com\u002Fp\u002Fa-visual-guide-to-mixture-of-experts](https:\u002F\u002Fnewsletter.maartengrootendorst.com\u002Fp\u002Fa-visual-guide-to-mixture-of-experts) |\n| LLM微调                 | 模型训练         | [https:\u002F\u002Fmagazine.sebastianraschka.com\u002Fp\u002Ffinetuning-large-language-models](https:\u002F\u002Fmagazine.sebastianraschka.com\u002Fp\u002Ffinetuning-large-language-models)                   |\n| Transformer LLM的工作原理 | 架构               | [https:\u002F\u002Fwww.deeplearning.ai\u002Fshort-courses\u002Fhow-transformer-llms-work\u002F](https:\u002F\u002Fwww.deeplearning.ai\u002Fshort-courses\u002Fhow-transformer-llms-work\u002F)                           |\n| 从零开始构建GPT           | 实践操作           | [https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=kCc8FmEb1nY](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=kCc8FmEb1nY)                                                                             |\n| LLM课程（GitHub）         | 结构化学习    | [https:\u002F\u002Fgithub.com\u002Fmlabonne\u002Fllm-course](https:\u002F\u002Fgithub.com\u002Fmlabonne\u002Fllm-course)                                                                                       |\n| LLM课程（Hugging Face）   | 实用LLM          | [https:\u002F\u002Fhuggingface.co\u002Flearn\u002Fllm-course\u002Fchapter1\u002F1](https:\u002F\u002Fhuggingface.co\u002Flearn\u002Fllm-course\u002Fchapter1\u002F1)                                                               |\n| 优秀的LLM应用            | 项目创意          | [https:\u002F\u002Fgithub.com\u002FShubhamsaboo\u002Fawesome-llm-apps](https:\u002F\u002Fgithub.com\u002FShubhamsaboo\u002Fawesome-llm-apps)                                                                   |\n| RAG如何增强LLM           | RAG                | [https:\u002F\u002Fawesomeneuron.substack.com\u002Fp\u002Fhow-rag-enhances-llms-a-step-by-step](https:\u002F\u002Fawesomeneuron.substack.com\u002Fp\u002Fhow-rag-enhances-llms-a-step-by-step)                                                       |\n| AI代理视觉指南          | AI代理              | [https:\u002F\u002Fawesomeneuron.substack.com\u002Fp\u002Fa-visual-guide-to-ai-agents](https:\u002F\u002Fawesomeneuron.substack.com\u002Fp\u002Fa-visual-guide-to-ai-agents)                                        |\n\n## 贡献说明\n\n欢迎各位贡献！\n\n请确保：\n\n- 笔记本整洁（提交前重启并运行所有单元）\n- 遵循现有结构与命名规范\n- PR内容聚焦、易读且有文档说明\n- 在RAG、OCR等文件夹中，请保持已清理的结构部分\n- 如需新增文件夹，请按照规范结构创建。\n\n## 许可证\n\n- 本项目采用MIT许可证。详情请参阅`LICENSE`文件。\n\n## 与我联系\n\n[![X](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FX-000000?style=for-the-badge&logo=x&logoColor=white)](https:\u002F\u002Fx.com\u002Ftechwith_ram)\n[![Instagram](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FInstagram-E4405F?style=for-the-badge&logo=instagram&logoColor=white)](https:\u002F\u002Finstagram.com\u002Ftechwith.ram)\n[![GitHub](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FGitHub-181717?style=for-the-badge&logo=github&logoColor=white)](https:\u002F\u002Fgithub.com\u002FRamakm)","# ai-hands-on 快速上手指南\n\n本指南旨在帮助开发者从零开始掌握 AI 工程化技能，涵盖数学基础、PyTorch、神经网络、Transformer、RAG 及 OCR 等核心内容。\n\n## 环境准备\n\n在开始之前，请确保您的开发环境满足以下要求：\n\n- **操作系统**：Windows, macOS 或 Linux\n- **Python 版本**：推荐 Python 3.8 及以上版本\n- **包管理器**：pip\n- **运行环境**：推荐安装 Jupyter Lab 或 Jupyter Notebook 以交互式运行教程笔记\n\n## 安装步骤\n\n1. **克隆仓库**\n   将项目代码下载到本地：\n   ```bash\n   git clone https:\u002F\u002Fgithub.com\u002FRamakm\u002Fai-hands-on.git\n   cd ai-hands-on\n   ```\n\n2. **安装核心依赖**\n   安装项目主目录所需的基础依赖包。\n   *注：国内用户可使用清华源或阿里源加速安装。*\n   ```bash\n   pip install -r requirements.txt -i https:\u002F\u002Fpypi.tuna.tsinghua.edu.cn\u002Fsimple\n   ```\n\n3. **安装子模块依赖（可选）**\n   如果您计划学习 RAG 或 OCR 章节，需分别进入对应文件夹安装额外依赖：\n   ```bash\n   # 安装 RAG 模块依赖\n   cd 5.RAG\n   pip install -r requirements.txt -i https:\u002F\u002Fpypi.tuna.tsinghua.edu.cn\u002Fsimple\n   cd ..\n\n   # 安装 OCR 模块依赖\n   cd 6.OCR\n   pip install -r requirements.txt -i https:\u002F\u002Fpypi.tuna.tsinghua.edu.cn\u002Fsimple\n   cd ..\n   ```\n\n4. **启动 Jupyter 环境**\n   在项目根目录启动 Jupyter Lab：\n   ```bash\n   jupyter lab\n   # 或者使用经典笔记本界面\n   # jupyter notebook\n   ```\n\n## 基本使用\n\n启动 Jupyter 后，浏览器会自动打开项目目录。请按照以下推荐路径进行学习实践：\n\n### 1. 核心学习路径（按顺序执行）\n依次打开并运行以下文件夹中的 `.ipynb` 文件，构建从数学到 Transformer 的完整知识体系：\n- `1.Math\u002F`：数学基础（微积分、线性代数、概率统计）\n- `2.PyTorch\u002F`：PyTorch 张量操作与基础\n- `3.Neural-Network(NN)\u002F`：从零构建神经网络、激活函数与优化器\n- `4.Transformer\u002F`：注意力机制与 Decoder-only 架构\n\n### 2. 专项实战模块（独立运行）\n完成核心路径后，可单独探索以下高级应用：\n- `5.RAG\u002F`：检索增强生成全流程（索引、检索、向量数据库集成）\n- `6.OCR\u002F`：光学字符识别管道与图像预处理\n\n### 3. 传统机器学习回顾\n如需复习基础模型，可参考以下实现：\n- `1.Linear Regression`\n- `2.Logistic Regression`\n- `3.Decision Tree Model`\n- `4.Naive Bayes Classification`\n\n> **提示**：在提交代码或深入新章节前，建议在 Jupyter 中选择 \"Kernel\" -> \"Restart & Run All\" 以确保笔记运行干净无误。","一位刚转行 AI 的工程师试图从零构建一个基于 RAG 的企业文档问答系统，却在数学原理和代码实现的断层中举步维艰。\n\n### 没有 ai-hands-on 时\n- **理论脱节**：在各类教程中碎片化学习线性代数和概率论，无法理解这些数学公式如何转化为具体的 PyTorch 张量操作。\n- **黑盒困境**：直接调用 Hugging Face 现成模型，一旦遇到注意力机制报错或梯度消失，因不懂底层架构而束手无策。\n- **路径迷茫**：面对海量的 RAG、OCR 和 Transformer 资料，缺乏清晰的进阶路线，花费数周仍在环境配置和基础概念间打转。\n- **实战缺失**：看过无数理论视频，却从未亲手从零写过一层神经网络或完整的检索增强生成流水线，面试与实战均缺乏底气。\n\n### 使用 ai-hands-on 后\n- **知行合一**：通过 `Math Fundamentals` 到 `PyTorch Basics` 的连贯笔记，直观看到导数和矩阵运算如何直接驱动代码中的张量变换。\n- **透视内核**：跟随 `Neural-Network` 和 `Transformers` 章节，亲手从零搭建神经元和解码器架构，彻底搞懂 Self-Attention 的内部运作逻辑。\n- **结构化进阶**：依据推荐的 Learning Path，按部就班地从基础数学过渡到复杂的 RAG 管道构建，大幅缩短摸索时间。\n- **端到端落地**：利用 `RAG` 和 `OCR` 专项实战笔记本，快速跑通从图像预处理、文本提取到向量检索的全流程，成功交付可运行的原型系统。\n\nai-hands-on 将抽象的 AI 工程知识拆解为可执行的代码步骤，帮助开发者从“调包侠”蜕变为能从头构建系统的真正 AI 工程师。","https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FRamakm_ai-hands-on_9e302bc8.png","Ramakm","Ramakrushna Mohapatra","https:\u002F\u002Foss.gittoolsai.com\u002Favatars\u002FRamakm_c4df7df3.png","Sr. Data Scientist. Gen AI & LLM Whisper . Tech Trainer . Founder","www.growtechie.in","Bengaluru, Ind","itsramakrushna@gmail.com","techwith_ram","ramakrushna.tech","https:\u002F\u002Fgithub.com\u002FRamakm",[86,90],{"name":87,"color":88,"percentage":89},"Jupyter Notebook","#DA5B0B",99.1,{"name":91,"color":92,"percentage":93},"Python","#3572A5",0.9,1078,246,"2026-04-05T15:59:26","MIT","未说明",{"notes":100,"python":98,"dependencies":101},"项目包含多个子模块（如 RAG 和 OCR），这些子文件夹内可能有独立的 requirements.txt 文件，需分别安装额外依赖。建议使用 Jupyter Lab 或 Jupyter Notebook 运行提供的笔记本文件进行学习。",[102,103,104,105,106,107,108,109,110],"torch","transformers","scikit-learn","xgboost","lightgbm","catboost","jupyter","notebook","lab",[26,13,15,14],[113,114,115,116,117,118,119,120,121,122,123,124,125],"ai","artificial-intelligence","books","chatbot","machine-learning","math","ml","mlmodel","neural-network","ocr","pytorch","rag","transformer",null,"2026-03-27T02:49:30.150509","2026-04-06T14:05:17.567014",[],[]]