[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"similar-probml--pml-book":3,"tool-probml--pml-book":65},[4,23,32,40,49,57],{"id":5,"name":6,"github_repo":7,"description_zh":8,"stars":9,"difficulty_score":10,"last_commit_at":11,"category_tags":12,"status":22},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。",84991,2,"2026-04-05T10:45:23",[13,14,15,16,17,18,19,20,21],"图像","数据工具","视频","插件","Agent","其他","语言模型","开发框架","音频","ready",{"id":24,"name":25,"github_repo":26,"description_zh":27,"stars":28,"difficulty_score":29,"last_commit_at":30,"category_tags":31,"status":22},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。",77062,3,"2026-04-04T04:44:48",[17,13,20,19,18],{"id":33,"name":34,"github_repo":35,"description_zh":36,"stars":37,"difficulty_score":29,"last_commit_at":38,"category_tags":39,"status":22},519,"PaddleOCR","PaddlePaddle\u002FPaddleOCR","PaddleOCR 是一款基于百度飞桨框架开发的高性能开源光学字符识别工具包。它的核心能力是将图片、PDF 等文档中的文字提取出来，转换成计算机可读取的结构化数据，让机器真正“看懂”图文内容。\n\n面对海量纸质或电子文档，PaddleOCR 解决了人工录入效率低、数字化成本高的问题。尤其在人工智能领域，它扮演着连接图像与大型语言模型（LLM）的桥梁角色，能将视觉信息直接转化为文本输入，助力智能问答、文档分析等应用场景落地。\n\nPaddleOCR 适合开发者、算法研究人员以及有文档自动化需求的普通用户。其技术优势十分明显：不仅支持全球 100 多种语言的识别，还能在 Windows、Linux、macOS 等多个系统上运行，并灵活适配 CPU、GPU、NPU 等各类硬件。作为一个轻量级且社区活跃的开源项目，PaddleOCR 既能满足快速集成的需求，也能支撑前沿的视觉语言研究，是处理文字识别任务的理想选择。",74939,"2026-04-05T23:16:38",[19,13,20,18],{"id":41,"name":42,"github_repo":43,"description_zh":44,"stars":45,"difficulty_score":46,"last_commit_at":47,"category_tags":48,"status":22},3215,"awesome-machine-learning","josephmisiti\u002Fawesome-machine-learning","awesome-machine-learning 是一份精心整理的机器学习资源清单，汇集了全球优秀的机器学习框架、库和软件工具。面对机器学习领域技术迭代快、资源分散且难以甄选的痛点，这份清单按编程语言（如 Python、C++、Go 等）和应用场景（如计算机视觉、自然语言处理、深度学习等）进行了系统化分类，帮助使用者快速定位高质量项目。\n\n它特别适合开发者、数据科学家及研究人员使用。无论是初学者寻找入门库，还是资深工程师对比不同语言的技术选型，都能从中获得极具价值的参考。此外，清单还延伸提供了免费书籍、在线课程、行业会议、技术博客及线下聚会等丰富资源，构建了从学习到实践的全链路支持体系。\n\n其独特亮点在于严格的维护标准：明确标记已停止维护或长期未更新的项目，确保推荐内容的时效性与可靠性。作为机器学习领域的“导航图”，awesome-machine-learning 以开源协作的方式持续更新，旨在降低技术探索门槛，让每一位从业者都能高效地站在巨人的肩膀上创新。",72149,1,"2026-04-03T21:50:24",[20,18],{"id":50,"name":51,"github_repo":52,"description_zh":53,"stars":54,"difficulty_score":46,"last_commit_at":55,"category_tags":56,"status":22},2234,"scikit-learn","scikit-learn\u002Fscikit-learn","scikit-learn 是一个基于 Python 构建的开源机器学习库，依托于 SciPy、NumPy 等科学计算生态，旨在让机器学习变得简单高效。它提供了一套统一且简洁的接口，涵盖了从数据预处理、特征工程到模型训练、评估及选择的全流程工具，内置了包括线性回归、支持向量机、随机森林、聚类等在内的丰富经典算法。\n\n对于希望快速验证想法或构建原型的数据科学家、研究人员以及 Python 开发者而言，scikit-learn 是不可或缺的基础设施。它有效解决了机器学习入门门槛高、算法实现复杂以及不同模型间调用方式不统一的痛点，让用户无需重复造轮子，只需几行代码即可调用成熟的算法解决分类、回归、聚类等实际问题。\n\n其核心技术亮点在于高度一致的 API 设计风格，所有估算器（Estimator）均遵循相同的调用逻辑，极大地降低了学习成本并提升了代码的可读性与可维护性。此外，它还提供了强大的模型选择与评估工具，如交叉验证和网格搜索，帮助用户系统地优化模型性能。作为一个由全球志愿者共同维护的成熟项目，scikit-learn 以其稳定性、详尽的文档和活跃的社区支持，成为连接理论学习与工业级应用的最",65628,"2026-04-05T10:10:46",[20,18,14],{"id":58,"name":59,"github_repo":60,"description_zh":61,"stars":62,"difficulty_score":10,"last_commit_at":63,"category_tags":64,"status":22},3364,"keras","keras-team\u002Fkeras","Keras 是一个专为人类设计的深度学习框架，旨在让构建和训练神经网络变得简单直观。它解决了开发者在不同深度学习后端之间切换困难、模型开发效率低以及难以兼顾调试便捷性与运行性能的痛点。\n\n无论是刚入门的学生、专注算法的研究人员，还是需要快速落地产品的工程师，都能通过 Keras 轻松上手。它支持计算机视觉、自然语言处理、音频分析及时间序列预测等多种任务。\n\nKeras 3 的核心亮点在于其独特的“多后端”架构。用户只需编写一套代码，即可灵活选择 TensorFlow、JAX、PyTorch 或 OpenVINO 作为底层运行引擎。这一特性不仅保留了 Keras 一贯的高层易用性，还允许开发者根据需求自由选择：利用 JAX 或 PyTorch 的即时执行模式进行高效调试，或切换至速度最快的后端以获得最高 350% 的性能提升。此外，Keras 具备强大的扩展能力，能无缝从本地笔记本电脑扩展至大规模 GPU 或 TPU 集群，是连接原型开发与生产部署的理想桥梁。",63927,"2026-04-04T15:24:37",[20,14,18],{"id":66,"github_repo":67,"name":68,"description_en":69,"description_zh":70,"ai_summary_zh":70,"readme_en":71,"readme_zh":72,"quickstart_zh":73,"use_case_zh":74,"hero_image_url":75,"owner_login":76,"owner_name":77,"owner_avatar_url":78,"owner_bio":79,"owner_company":80,"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":46,"env_os":98,"env_gpu":98,"env_ram":98,"env_deps":99,"category_tags":102,"github_topics":80,"view_count":103,"oss_zip_url":80,"oss_zip_packed_at":80,"status":22,"created_at":104,"updated_at":105,"faqs":106,"releases":136},546,"probml\u002Fpml-book","pml-book","\"Probabilistic Machine Learning\" - a book series by Kevin Murphy","pml-book 是一个开源项目，托管了凯文·墨菲教授《概率机器学习》系列书籍的电子版及配套代码。它汇集了 Book 0、Book 1 和 Book 2 三个版本，为学习者提供免费的在线阅读渠道。针对机器学习理论教材昂贵、版本更新慢以及代码实现分散的问题，pml-book 提供了一个集中且权威的解决方案，确保大家能接触到最前沿的概率建模知识。\n\npml-book 主要面向人工智能领域的学生、研究人员以及希望夯实算法基础的开发者。其核心价值在于将复杂的数学理论与 Python 实践紧密结合，书中每一章都配有可运行的代码示例，帮助读者跨越从公式到实现的鸿沟。无论是入门级的监督学习，还是高阶的变分推断与生成模型，pml-book 都能提供系统化的指引。借助 pml-book，用户能够以零成本获取顶尖学术资源，高效构建概率机器学习思维框架，是深入理解 AI 原理的优质途径。","\n# \"Probabilistic machine learning\": a book series by Kevin Murphy\n\n\n  \u003Cp> \u003C\u002Fp>\n  \n## Book 0: \"Machine Learning: A Probabilistic Perspective\" (2012)\n\nSee [this link](https:\u002F\u002Fprobml.github.io\u002Fpml-book\u002Fbook0.html)\n\n\u003C!--\nSee [this link](https:\u002F\u002Fprobml.github.io\u002Fpml-book\u002Fpml0\u002Fbook0.html)\n-->\n\n## Book 1: \"Probabilistic Machine Learning: An Introduction\" (2022)\n\nSee [this link](https:\u002F\u002Fprobml.github.io\u002Fpml-book\u002Fbook1.html)\n\n\n## Book 2: \"Probabilistic Machine Learning: Advanced Topics\" (2023)\n\nSee [this link](https:\u002F\u002Fprobml.github.io\u002Fpml-book\u002Fbook2.html)\n\n\n\n","# “概率机器学习”：Kevin Murphy 所著的一套书籍系列\n\n\n  \u003Cp> \u003C\u002Fp>\n  \n## 第 0 卷：《机器学习：一种概率视角》（2012）\n\n参见 [此链接](https:\u002F\u002Fprobml.github.io\u002Fpml-book\u002Fbook0.html)\n\n\u003C!--\n参见 [此链接](https:\u002F\u002Fprobml.github.io\u002Fpml-book\u002Fpml0\u002Fbook0.html)\n-->\n\n## 第 1 卷：《概率机器学习：导论》（2022）\n\n参见 [此链接](https:\u002F\u002Fprobml.github.io\u002Fpml-book\u002Fbook1.html)\n\n\n## 第 2 卷：《概率机器学习：高级主题》（2023）\n\n参见 [此链接](https:\u002F\u002Fprobml.github.io\u002Fpml-book\u002Fbook2.html)","# pml-book 快速上手指南\n\n## 简介\n`pml-book` 是 Kevin Murphy 所著《概率机器学习》（Probabilistic Machine Learning）系列书籍的开源资源仓库。该项目提供了从基础到进阶的完整教材内容，适合机器学习从业者及研究者阅读。\n\n## 环境准备\n- **操作系统**: Windows \u002F macOS \u002F Linux\n- **运行环境**: 现代网页浏览器 (Chrome, Firefox, Edge 等)\n- **开发工具**: Git (仅当需要本地克隆时使用)\n- **网络要求**: 需访问 GitHub 及相关静态页面服务\n\n## 安装步骤\n### 方式一：在线访问（推荐）\n无需安装任何软件，直接通过浏览器访问以下官方链接即可阅读：\n\n- **Book 0**: https:\u002F\u002Fprobml.github.io\u002Fpml-book\u002Fbook0.html\n- **Book 1**: https:\u002F\u002Fprobml.github.io\u002Fpml-book\u002Fbook1.html\n- **Book 2**: https:\u002F\u002Fprobml.github.io\u002Fpml-book\u002Fbook2.html\n\n### 方式二：本地克隆\n如需离线查阅或研究源码结构，可使用 Git 克隆仓库：\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002Fprobml\u002Fpml-book.git\n```\n\n> **注意**：由于托管于 GitHub，国内用户访问可能较慢，建议使用网络加速工具或 GitHub 镜像源。\n\n## 基本使用\n1. **在线阅读**：点击上述任意链接，直接在浏览器中浏览书籍章节。\n2. **本地阅读**：克隆仓库后，进入目录查找对应的 HTML 文件或 PDF 资源进行离线阅读。\n3. **版本选择**：\n   - 初学者建议从 **Book 0** 开始。\n   - 已掌握基础并寻求进阶内容的用户可选择 **Book 1** 和 **Book 2**。","某资深算法工程师在构建医疗诊断模型时，需要引入不确定性量化模块，但面对复杂的概率推断理论感到无从下手。\n\n### 没有 pml-book 时\n- 知识点散落在各类学术论文中，难以形成系统的认知框架\n- 不同文献对贝叶斯网络符号定义冲突，导致公式推导频繁出错\n- 理论讲解过于抽象，缺乏结合 PyTorch 或 TensorFlow 的代码验证\n- 遇到变分自编码器等高级话题时，需反复查阅多本教材才能理清脉络\n\n### 使用 pml-book 后\n- pml-book 按难度分级组织内容，快速定位从入门到精通的学习路径\n- 全书保持统一的数学符号体系，大幅降低阅读时的认知负荷\n- 书中章节直接关联官方代码库，能立即复现并调试概率模型\n- 针对高级主题提供详细推导过程，帮助快速攻克变分推断等技术瓶颈\n\npml-book 通过整合权威理论与工程实践，让概率机器学习知识的掌握变得高效且直观。","https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fprobml_pml-book_2db4dc92.png","probml","Probabilistic machine learning","https:\u002F\u002Foss.gittoolsai.com\u002Favatars\u002Fprobml_f9a37f2f.jpg","Material to accompany my book series \"Probabilistic Machine Learning\" (Software, Data, Exercises, Figures, etc)",null,"murphyk@gmail.com","sirbayes","probml.ai","https:\u002F\u002Fgithub.com\u002Fprobml",[86,90],{"name":87,"color":88,"percentage":89},"Jupyter Notebook","#DA5B0B",95.6,{"name":91,"color":92,"percentage":93},"HTML","#e34c26",4.4,5542,642,"2026-04-04T21:26:06","MIT","未说明",{"notes":100,"python":98,"dependencies":101},"该仓库主要提供 Kevin Murphy 概率机器学习系列书籍的在线链接，README 中未提及具体的代码运行环境或依赖信息。",[],[18],8,"2026-03-27T02:49:30.150509","2026-04-06T09:46:11.343055",[107,112,116,121,126,131],{"id":108,"question_zh":109,"answer_zh":110,"source_url":111},2205,"如何确认我拥有的 PML 书籍版本对应的发布日期？","可以访问 GitHub 发布页面查看版本信息：https:\u002F\u002Fgithub.com\u002Fprobml\u002Fpml-book\u002Freleases","https:\u002F\u002Fgithub.com\u002Fprobml\u002Fpml-book\u002Fissues\u002F9",{"id":113,"question_zh":114,"answer_zh":115,"source_url":111},2206,"如何确认书中的错误是否已在最新版本中修复？","维护者会在评论中确认修复状态（如 'all fixed in latest release'），建议检查 GitHub Releases 页面的最新发行版以获取修正后的内容。",{"id":117,"question_zh":118,"answer_zh":119,"source_url":120},2207,"书中是否包含已知的拼写或公式错误？","是的，GitHub Issues 中记录了多轮勘误（涉及第 2-9 章及附录），维护者 murphyk 表示已在最新版本中修复了所有这些问题。","https:\u002F\u002Fgithub.com\u002Fprobml\u002Fpml-book\u002Fissues\u002F40",{"id":122,"question_zh":123,"answer_zh":124,"source_url":125},2208,"关于“完全类”定理归属 Wald 的争议，作者最终如何处理？","作者修正了拼写错误（admissible），但保留了原文内容，并引用 Dan 和 Jaynes 的文献供读者自行判断。","https:\u002F\u002Fgithub.com\u002Fprobml\u002Fpml-book\u002Fissues\u002F210",{"id":127,"question_zh":128,"answer_zh":129,"source_url":130},2209,"第 8 章优化部分的拉格朗日函数符号是否有修正？","是的，将拉格朗日符号改为常规 L，并将 mu 和 lambda 的系数符号改为正号 (+)，以避免与集合符号冲突并符合标准。","https:\u002F\u002Fgithub.com\u002Fprobml\u002Fpml-book\u002Fissues\u002F108",{"id":132,"question_zh":133,"answer_zh":134,"source_url":135},2210,"如何跟踪草稿版本中的小错误而不单独发起 Issue？","维护者创建了专门的 Issue（如 #119）来追踪不需要单独 Issue 的小错误，用户可直接在该 Issue 下留言反馈，维护者会统一处理。","https:\u002F\u002Fgithub.com\u002Fprobml\u002Fpml-book\u002Fissues\u002F119",[137,142,147,152,157,162,167,172,177,182,187,192,197,202,207,212,217,222,227,232],{"id":138,"version":139,"summary_zh":140,"released_at":141},111395,"2022-07-29","- Fixed numerous typos.\r\n- Major change in the way we refer to code.\r\nI used to use to figure captions like `Generated by code.probml.ai\u002Fbook1\u002F1.3`.\r\nNow I just give the filename `Generated by iris_plot.ipynb` which refer to strings in the lookup table at \r\nhttps:\u002F\u002Fprobml.github.io\u002Fnotebooks\r\n\r\nSee before and after:\r\n\u003Cimg width=\"1515\" alt=\"Screen Shot 2022-07-29 at 11 55 51 AM\" src=\"https:\u002F\u002Fuser-images.githubusercontent.com\u002F4632336\u002F181827444-db084b3e-5526-43d8-9c8f-a27ff2100751.png\">\r\n","2022-07-29T19:10:23",{"id":143,"version":144,"summary_zh":145,"released_at":146},111396,"2022-05-09","- change name of downloadable asset from pml1.pdf to book1.pdf for consistency\r\n-  fixed a few typos (see https:\u002F\u002Fgithub.com\u002Fprobml\u002Fpml-book\u002Fissues\u002F361)","2022-05-09T18:15:11",{"id":148,"version":149,"summary_zh":150,"released_at":151},111397,"2022-05-05","- fixed a few minor typos listed here: https:\u002F\u002Fgithub.com\u002Fprobml\u002Fpml-book\u002Fissues\u002F354\r\n- major change to the figure-generating code, details at https:\u002F\u002Fgithub.com\u002Fprobml\u002Fpml-book\u002Fissues\u002F355. The only change to the book itself is that figures.probml.ai URLs are reanmed to code.probml.ai, and these now point to individual notebooks, instead of one per chapter.\r\n- (Lots of changes to the backend that won't affect the reader.)","2022-05-06T04:48:33",{"id":153,"version":154,"summary_zh":155,"released_at":156},111387,"2025-04-18","- fixed all typos to date\r\n- tweaked sec 15.7 on LLMs to clarify difference between non-generative LMs (eg BERT) and generative LMs (eg GPT). \r\n- Added some references to recent LLM books (eg Burkov's \"100 page LLM\" book, Lambert's \"RLHF\" book, Narayanan's \"AI Snake Oil\" book)","2025-04-18T10:49:50",{"id":158,"version":159,"summary_zh":160,"released_at":161},111388,"2024-11-23","Fixed all github issues. Sending to MIT Press for third print run.","2024-11-23T19:03:26",{"id":163,"version":164,"summary_zh":165,"released_at":166},111389,"2024-06-26","- Fixed 2 new github issues that arose since 2024-06-12 release (one of them an important math error in the Jacobian \r\nderivation for the activation functions in sec 13.4.3)","2024-06-27T04:41:45",{"id":168,"version":169,"summary_zh":170,"released_at":171},111390,"2024-06-12","Fixed all github issues filed since June 2023!\r\nThe book is now 2 pages longer, due to addition of a few drops of extra content here and there (eg sec 5.2.5.3 on MDL\r\nand 13.5.7 on double descent).","2024-06-13T04:54:02",{"id":173,"version":174,"summary_zh":175,"released_at":176},111391,"2023-06-21","For the 2023-06-20 I uploaded the wrong file. This version corrects that mistake.\r\n(Thanks to Umer Javed for spotting this :)\r\n\r\n","2023-06-22T15:32:50",{"id":178,"version":179,"summary_zh":180,"released_at":181},111392,"2023-06-20","Fixed a few typos, clarified writing in a few sections, added a few more references. (See github issues for details.)","2023-06-20T22:50:09",{"id":183,"version":184,"summary_zh":185,"released_at":186},111393,"2023-04-17","Many small changes for the 2nd printing.\r\nComplete change list at \r\n https:\u002F\u002Fgithub.com\u002Fprobml\u002Fpml-book\u002Fissues?page=2&q=is%3Aissue+created%3A2022-08-08..2023-04-17\r\n\r\nChanged figures in 2nd printing\r\nfig 6.2 p204\r\nfig 7.6  p252\r\nfig 9.2 p320\r\nfig 10.5 p341\r\nfig 18.1 p600\r\n\r\n\r\n","2023-04-17T21:58:42",{"id":188,"version":189,"summary_zh":190,"released_at":191},111394,"2022-08-08","fix a few typos\r\n\r\n","2022-08-09T05:50:02",{"id":193,"version":194,"summary_zh":195,"released_at":196},111398,"2022-04-09","Fixed various typos (details are at https:\u002F\u002Fgithub.com\u002Fprobml\u002Fpml-book\u002Fissues\u002F304).\r\nNote that the page numbering is slightly different from previous versions, but other numbering (section, equation, figure) is the same.\r\n\r\nAlso, I am working with Zeel Patel and Nipun Batra to change the code for all the figures: in the new regime, each figure will have its own colab (stored [here](https:\u002F\u002Fgithub.com\u002Fprobml\u002Fpyprobml\u002Ftree\u002Fmaster\u002Fnotebooks\u002Fbook1)),Also the figures will be [\"latexified\"](https:\u002F\u002Fnipunbatra.github.io\u002Fblog\u002Fvisualisation\u002F2014\u002F06\u002F02\u002Flatexify.html) to look nicer. \r\n(In the past, each figure has its own script, which is then imported into one auto-generated colab per chapter stored [here](https:\u002F\u002Fgithub.com\u002Fprobml\u002Fpml-book\u002Ftree\u002Fmain\u002Fpml1\u002Ffigure_notebooks).)\r\nSo far just figs 2.5, 2.6 and 3.3 have been changed to this new format. One consequence of this is that clicking on the link to, say, figures.probml.ai\u002Fbook1\u002F3.3 will take you to [this cell in the old colab](colab.research.google.com\u002Fgithub\u002Fprobml\u002Fpml-book\u002Fblob\u002Fmain\u002Fpml1\u002Ffigure_notebooks\u002Fchapter3_probability_multivariate_models_figures.ipynb#scrollTo=CTcC3CxkxqZZ), which points to the old code\u002Ffigure, and note the new one (shown below). Consequently there is a small discrepancy between the appearance of some figures and the print edition. This will be fixed in a future release. \r\n\r\n\u003Cimg width=\"676\" alt=\"Screen Shot 2022-04-09 at 5 01 54 PM\" src=\"https:\u002F\u002Fuser-images.githubusercontent.com\u002F4632336\u002F162595628-2b066cb0-da65-435b-9f40-06e1b7a659fd.png\">\r\n\r\n","2022-04-10T00:16:31",{"id":198,"version":199,"summary_zh":200,"released_at":201},111399,"2022-02-08","This is the same as 2022-02-07 except for the equation numbers in sec 7.7.3 (solving underconstrained systems (least norm estimation)).\r\nThis section omits some erroneous material from an earlier version, so I have modified the numbering at the end of this section so remaining equation numbers match previous versions of the book.","2022-02-09T02:34:38",{"id":203,"version":204,"summary_zh":205,"released_at":206},111400,"2022-02-07","2022-02-07 release","2022-02-08T03:40:17",{"id":208,"version":209,"summary_zh":210,"released_at":211},111401,"2021-11-08","Minor updates to the version sent to MIT Press. Closes all issues to date.\r\nDetails of changes are listed at https:\u002F\u002Fgithub.com\u002Fprobml\u002Fpml-book\u002Fissues\u002F247\r\n","2021-11-08T18:38:14",{"id":213,"version":214,"summary_zh":215,"released_at":216},111402,"camera-2021-08-30","This is the version I sent to MIT Press for the first printing :)\r\n","2021-08-30T21:41:59",{"id":218,"version":219,"summary_zh":220,"released_at":221},111403,"2021-08-03","- fixed many typos\r\n- tweaked many figures to make CMYK version look nicer","2021-08-03T05:58:23",{"id":223,"version":224,"summary_zh":225,"released_at":226},111404,"2021-06-05","- moved functional AD to vol 2\r\n- moved adversarial examples to vol 2 (since not specific to CNN chapter and no other home for it)\r\n -fixed typos from github\r\n","2021-06-06T04:30:38",{"id":228,"version":229,"summary_zh":230,"released_at":231},111405,"2021-05-21","- fixed numerous typos (see closed github issues)\r\n- fixed eqns 11.128-11.29 for weight posterior in Bayesian linear regression (surprised this one slipped through the eyes of 52k readers!)\r\n- added short section on weakly supervised learning to the very end of ch. 19\r\n- moved the section on information criterion  (5.1.7) to after the section on model selection (5.2.2)\r\n- removed section on probability integral transform (2.8.8) since it will be covered in vol 2","2021-05-22T06:33:12",{"id":233,"version":234,"summary_zh":235,"released_at":236},111406,"2021-05-06","This is the version I sent to MIT Press for \"final\" copy editing.","2021-05-06T19:21:04"]