[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"similar-PRML--PRMLT":3,"tool-PRML--PRMLT":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",138956,2,"2026-04-05T11:33:21",[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":75,"owner_avatar_url":76,"owner_bio":77,"owner_company":78,"owner_location":78,"owner_email":79,"owner_twitter":78,"owner_website":80,"owner_url":81,"languages":82,"stars":87,"forks":88,"last_commit_at":89,"license":90,"difficulty_score":23,"env_os":91,"env_gpu":92,"env_ram":92,"env_deps":93,"category_tags":100,"github_topics":101,"view_count":23,"oss_zip_url":78,"oss_zip_packed_at":78,"status":16,"created_at":107,"updated_at":108,"faqs":109,"releases":150},3336,"PRML\u002FPRMLT","PRMLT","Matlab code of machine learning algorithms in book PRML","PRMLT 是一套基于 MATLAB 编写的机器学习算法代码库，完整复现了经典教材《模式识别与机器学习》（PRML）中的核心内容。它主要解决了学习者在研读理论时难以将数学公式转化为实际代码的痛点，帮助用户直观地验证和理解复杂的算法原理。\n\n这套工具特别适合高校研究人员、数据科学学生以及需要深入探究算法底层的开发者使用。无论是用于教学演示、学术研究，还是作为自定义算法的开发基石，PRMLT 都能提供极大的便利。\n\n在技术实现上，PRMLT 具有鲜明的亮点：代码极度精简且无外部依赖，核心逻辑一目了然；通过向量化运算和矩阵分解等技巧，其运行效率远超 MATLAB 内置函数；同时注重数值稳定性，采用对数域概率计算等方法确保结果可靠。此外，代码注释详尽，变量符号与原著严格对应，并标注了相关公式出处，极大地提升了可读性。只需简单的初始化和演示运行，用户即可快速上手，探索机器学习的奥秘。","Introduction\n-------\nThis Matlab package implements machine learning algorithms described in the great textbook:\nPattern Recognition and Machine Learning by C. Bishop ([PRML](http:\u002F\u002Fresearch.microsoft.com\u002Fen-us\u002Fum\u002Fpeople\u002Fcmbishop\u002Fprml\u002F)).\n\nIt is written purely in Matlab language. It is self-contained. There is no external dependency.\n\nNote: this package requires Matlab **R2016b** or latter, since it utilizes a new Matlab syntax called [Implicit expansion](https:\u002F\u002Fcn.mathworks.com\u002Fhelp\u002Fmatlab\u002Frelease-notes.html?rntext=implicit+expansion&startrelease=R2016b&endrelease=R2016b&groupby=release&sortby=descending) (a.k.a. broadcasting). It also requires Statistics Toolbox (for some simple random number generator) and Image Processing Toolbox (for reading image data).\n\nDesign Goal\n-------\n* Succinct: The code is extremely compact. Minimizing code length is a major goal. As a result, the core of the algorithms can be easily spotted.\n* Efficient: Many tricks for speeding up Matlab code are applied (e.g. vectorization, matrix factorization, etc.). Usually, functions in this package are orders faster than Matlab builtin ones (e.g. kmeans).\n* Robust: Many tricks for numerical stability are applied, such as computing probability in logrithm domain, square root matrix update to enforce matrix symmetry\\PD, etc.\n* Readable: The code is heavily commented. Corresponding formulas in PRML are annoted. Symbols are in sync with the book.\n* Practical: The package is not only readable, but also meant to be easily used and modified to facilitate ML research. Many functions in this package are already widely used (see [Matlab file exchange](http:\u002F\u002Fwww.mathworks.com\u002Fmatlabcentral\u002Ffileexchange\u002F?term=authorid%3A49739)).\n\nInstallation\n-------\n1. Download the package to a local folder (e.g. ~\u002FPRMLT\u002F) by running: \n```console\ngit clone https:\u002F\u002Fgithub.com\u002FPRML\u002FPRMLT.git\n```\n2. Run Matlab and navigate to the folder (~\u002FPRMLT\u002F), then run the init.m script.\n\n3. Run some demos in ~\u002FPRMLT\u002Fdemo folder. Enjoy!\n\nFeedBack\n-------\nIf you find any bug or have any suggestion, please do file issues. I am graceful for any feedback and will do my best to improve this package.\n\nLicense\n-------\nReleased under MIT license\n\nContact\n-------\nsth4nth at gmail dot com\n","引言\n-------\n本Matlab工具包实现了由C. Bishop所著的经典教材《模式识别与机器学习》中描述的机器学习算法（[PRML](http:\u002F\u002Fresearch.microsoft.com\u002Fen-us\u002Fum\u002Fpeople\u002Fcmbishop\u002Fprml\u002F)）。\n\n该工具包完全用Matlab语言编写，自包含，无外部依赖。\n\n注意：本工具包需要Matlab **R2016b** 或更高版本，因为它使用了名为[隐式扩展](https:\u002F\u002Fcn.mathworks.com\u002Fhelp\u002Fmatlab\u002Frelease-notes.html?rntext=implicit+expansion&startrelease=R2016b&endrelease=R2016b&groupby=release&sortby=descending)（又称广播）的新语法。此外，它还需要Statistics Toolbox（用于一些简单的随机数生成器）和Image Processing Toolbox（用于读取图像数据）。\n\n设计目标\n-------\n* 简洁：代码极其紧凑，最小化代码长度是主要目标之一。因此，算法的核心部分一目了然。\n* 高效：采用了多种加速Matlab代码的技巧（如向量化、矩阵分解等）。通常，本工具包中的函数比Matlab内置函数快几个数量级（例如kmeans）。\n* 鲁棒：应用了许多提高数值稳定性的技巧，例如在对数域中计算概率、通过平方根矩阵更新来确保矩阵的对称性和正定性等。\n* 易读：代码注释丰富，并标注了与PRML中对应的公式，符号也与书中保持一致。\n* 实用：该工具包不仅易于阅读，还便于使用和修改，以促进机器学习研究。本工具包中的许多函数已被广泛使用（参见[Matlab文件交换](http:\u002F\u002Fwww.mathworks.com\u002Fmatlabcentral\u002Ffileexchange\u002F?term=authorid%3A49739)）。\n\n安装\n-------\n1. 通过以下命令将工具包下载到本地文件夹（例如~\u002FPRMLT\u002F）：\n```console\ngit clone https:\u002F\u002Fgithub.com\u002FPRML\u002FPRMLT.git\n```\n2. 打开Matlab并导航至该文件夹（~\u002FPRMLT\u002F），然后运行init.m脚本。\n\n3. 在~\u002FPRMLT\u002Fdemo文件夹中运行一些示例程序，尽情体验吧！\n\n反馈\n-------\n如果您发现任何错误或有任何建议，请提交问题。我非常感谢您的反馈，并将尽最大努力改进本工具包。\n\n许可证\n-------\n采用MIT许可证发布\n\n联系方式\n-------\nsth4nth at gmail dot com","# PRMLT 快速上手指南\n\nPRMLT 是一个纯 Matlab 实现的机器学习算法包，完整复现了经典教材《模式识别与机器学习》（Pattern Recognition and Machine Learning, PRML）中的核心算法。该工具包代码紧凑、高效且注释详尽，适合学习与科研使用。\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 \u002F macOS \u002F Linux（支持 Matlab 的平台均可）\n*   **Matlab 版本**：**R2016b** 或更高版本\n    *   *原因*：项目依赖 Matlab R2016b 引入的“隐式扩展”（Implicit expansion\u002FBroadcasting）语法特性。\n*   **必需工具箱**：\n    *   **Statistics and Machine Learning Toolbox**：用于基础随机数生成。\n    *   **Image Processing Toolbox**：用于读取图像数据。\n\n## 安装步骤\n\n1.  **下载源码**\n    打开终端（Terminal 或 CMD），运行以下命令将仓库克隆到本地文件夹（例如 `~\u002FPRMLT\u002F`）：\n    ```console\n    git clone https:\u002F\u002Fgithub.com\u002FPRML\u002FPRMLT.git\n    ```\n    > **提示**：如果国内访问 GitHub 速度较慢，可尝试使用镜像源或手动下载 ZIP 包解压至目标目录。\n\n2.  **初始化环境**\n    启动 Matlab，将当前工作目录切换至刚才下载的文件夹（如 `~\u002FPRMLT\u002F`），然后在命令行运行初始化脚本：\n    ```matlab\n    init.m\n    ```\n\n3.  **验证安装**\n    进入 `demo` 文件夹运行任意示例脚本，确认无误即可开始使用：\n    ```matlab\n    cd demo\n    % 运行任意演示脚本，例如：\n    demo_kmeans\n    ```\n\n## 基本使用\n\nPRMLT 的设计目标是与教材公式符号保持一致，因此使用方式非常直观。以下以调用高斯混合模型（GMM）相关函数为例：\n\n1.  **加载数据**（假设已有数据矩阵 `X`）\n2.  **调用算法**\n    直接调用对应的函数文件，变量命名通常与书中公式一致。例如进行聚类：\n    ```matlab\n    % 假设 X 是 NxD 的数据矩阵，K 为聚类中心数量\n    K = 3;\n    [mu, Sigma, pi_val] = gmmEm(X, K);\n    ```\n3.  **查看文档**\n    由于代码包含大量注释并标注了对应的书本公式，建议在 Matlab 中直接打开 `.m` 文件阅读头部注释，或在使用函数前输入：\n    ```matlab\n    help gmmEm\n    ```\n\n现在您可以浏览 `demo` 目录下的更多示例，深入探索 PRMLT 提供的丰富算法。","某高校科研团队正在复现《模式识别与机器学习》（PRML）书中的贝叶斯分类算法，以验证其在医学影像诊断中的有效性。\n\n### 没有 PRMLT 时\n- **公式对齐困难**：研究人员需手动将书本数学公式转换为 Matlab 代码，变量命名混乱，难以确认代码逻辑是否与理论一致。\n- **运行效率低下**：自行编写的循环结构未进行向量化优化，处理高分辨率病理切片数据时耗时极长，甚至超过内置函数的数倍。\n- **数值稳定性差**：在计算高维概率密度时容易因浮点数精度问题导致下溢，频繁出现非对称矩阵报错，调试数值稳定性耗费大量精力。\n- **依赖管理繁琐**：为复现一个算法需拼凑多个网络开源片段，外部依赖冲突频发，环境配置复杂且难以移植。\n\n### 使用 PRMLT 后\n- **符号无缝对应**：PRMLT 的代码变量与书中公式符号完全同步且注释详尽，研究者可直接对照书本核心章节定位算法逻辑，复现效率提升显著。\n- **执行速度飞跃**：利用隐式扩展和矩阵分解等技巧，PRMLT 实现了高度向量化，处理相同规模影像数据的耗时从小时级缩短至分钟级。\n- **鲁棒性内建保障**：内置对数域概率计算和平方根矩阵更新机制，自动规避数值下溢并强制矩阵对称正定，确保模型在极端数据下依然稳定收敛。\n- **开箱即用体验**：作为无外部依赖的独立包，只需运行初始化脚本即可调用所有算法，让团队能专注于医学特征分析而非环境调试。\n\nPRMLT 通过将经典理论转化为高效、稳定且可读的代码，极大地降低了机器学习算法从书本到实际科研应用的门槛。","https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FPRML_PRMLT_03fcd8ed.png","PRML","https:\u002F\u002Foss.gittoolsai.com\u002Favatars\u002FPRML_dbaf0feb.jpg","Pattern Recognition and Machine Learning",null,"sth4nth@gmail.com","http:\u002F\u002Fprml.github.com","https:\u002F\u002Fgithub.com\u002FPRML",[83],{"name":84,"color":85,"percentage":86},"MATLAB","#e16737",100,6205,2140,"2026-04-02T08:30:44","MIT","Windows, macOS, Linux","未说明",{"notes":94,"python":95,"dependencies":96},"该工具纯由 MATLAB 编写，无外部依赖。必须使用 MATLAB R2016b 或更高版本，因为利用了‘隐式扩展’（Implicit expansion）语法。此外，运行部分功能需要安装 Statistics Toolbox（用于随机数生成）和 Image Processing Toolbox（用于读取图像数据）。","不适用",[97,98,99],"MATLAB R2016b 或更高版本","Statistics Toolbox","Image Processing Toolbox",[13],[102,103,104,105,106],"matlab","machine-learning","prml","algorithms","machine-learning-algorithms","2026-03-27T02:49:30.150509","2026-04-06T05:15:22.331952",[110,115,120,125,130,135,140,145],{"id":111,"question_zh":112,"answer_zh":113,"source_url":114},15312,"使用自定义数据运行二分类 RVM 时，binPlot 函数报错\"Index exceeds matrix dimensions\"怎么办？","这个问题通常由两个原因导致：1. `binPlot` 函数不支持稀疏模型，建议改用 `plotClass` 函数进行绘图；2. 二分类器的标签假设为 (0, 1)，如果你的数据标签是 (1, 2)，需要将其转换为 (0, 1) 格式。","https:\u002F\u002Fgithub.com\u002FPRML\u002FPRMLT\u002Fissues\u002F47",{"id":116,"question_zh":117,"answer_zh":118,"source_url":119},15313,"logitBinPred.m 输出的预测标签似乎反了（0 变 1，1 变 0），如何修复？","这是一个已确认的 Bug。修复方法是将计算概率的代码从 `p = exp(-log1pexp(w'*X))` 修改为 `p = exp(-log1pexp(-w'*X))`，即在 `log1pexp` 函数内部对 `w'*X` 取负号，以修正数值稳定性公式中的符号错误。","https:\u002F\u002Fgithub.com\u002FPRML\u002FPRMLT\u002Fissues\u002F36",{"id":121,"question_zh":122,"answer_zh":123,"source_url":124},15314,"代码中混合高斯变分推断公式 (10.63) 的实现 `v = v0 + nk` 是否与书中不符？","经核查，PRML 原书和 MLAPP 书中此处均写为 `v = v0 + nk + 1`，但根据共轭高斯分布的理论推导，后验参数应为 `v = v0 + n`。维护者认为这可能是原书的笔误，代码中的 `+1` 可能是有意为之或推导差异，但在实际应用中两者差异极小，仅相当于先验值 `v0` 的不同。","https:\u002F\u002Fgithub.com\u002FPRML\u002FPRMLT\u002Fissues\u002F35",{"id":126,"question_zh":127,"answer_zh":128,"source_url":129},15315,"第 5 章的神经网络示例代码是否缺少偏置项（biases）？","是的，原示例代码为了简洁省略了偏置项，但这不符合书中的推荐做法且可能影响性能。维护者已确认并修复了此问题，添加了偏置项以提升模型表现。","https:\u002F\u002Fgithub.com\u002FPRML\u002FPRMLT\u002Fissues\u002F48",{"id":131,"question_zh":132,"answer_zh":133,"source_url":134},15316,"运行第 6 章 knLin_demo.m 时报错\"Too many input arguments\"如何解决？","这是演示脚本中的参数传递错误。脚本中调用了三个输入参数，但对应的函数定义只接受两个。维护者已修复该问题，请拉取最新代码即可解决。","https:\u002F\u002Fgithub.com\u002FPRML\u002FPRMLT\u002Fissues\u002F50",{"id":136,"question_zh":137,"answer_zh":138,"source_url":139},15317,"学习线性动态系统 (LDS) 时 EM 算法不收敛或不稳定怎么办？","PRML 书中描述的 LDS EM 算法在没有良好初始化的情况下很难工作且数值不稳定。建议使用子空间方法 (subspace method) 进行初始化，该方法虽未在 PRML 书中提及但在 BRML 书中有介绍。维护者已在代码库中添加了基于子空间方法的初始化功能。","https:\u002F\u002Fgithub.com\u002FPRML\u002FPRMLT\u002Fissues\u002F49",{"id":141,"question_zh":142,"answer_zh":143,"source_url":144},15318,"该代码库支持 GNU Octave 吗？","大部分代码和演示可以在 Octave 上运行，但并非全部兼容。例如 `chapter03\u002FlinRnd.m` 中使用的 `randg` 函数在 Octave 中即使安装了 statistics 包也无法正常工作。维护者目前没有计划专门适配 Octave，但欢迎社区提交兼容性补丁。","https:\u002F\u002Fgithub.com\u002FPRML\u002FPRMLT\u002Fissues\u002F45",{"id":146,"question_zh":147,"answer_zh":148,"source_url":149},15311,"这个代码库对应的是哪本书？","该代码库对应的书籍是《Pattern Recognition and Machine Learning》（模式识别与机器学习），作者是 Christopher Bishop。","https:\u002F\u002Fgithub.com\u002FPRML\u002FPRMLT\u002Fissues\u002F52",[151,156,160,165,170,175,180,185,190,195,200,205,210,215,220,225,230,235,240,245],{"id":152,"version":153,"summary_zh":154,"released_at":155},89962,"v2.2","小幅调整","2020-01-04T10:56:13",{"id":157,"version":158,"summary_zh":154,"released_at":159},89963,"v2.1","2019-07-14T12:05:30",{"id":161,"version":162,"summary_zh":163,"released_at":164},89964,"v2.0.1","小修复","2019-01-24T08:09:20",{"id":166,"version":167,"summary_zh":168,"released_at":169},89965,"v2.0","最终发布。","2018-12-06T16:06:02",{"id":171,"version":172,"summary_zh":173,"released_at":174},89966,"v2.0rc","添加多层感知机分类，并优化LDS相关功能。  \n添加内容说明和许可证信息。\n\n这是最终版本的候选发布版。今后，我将不再向该软件包添加任何新内容，只进行错误修复。请将此版本视为该软件包的最终状态。","2018-12-03T17:04:02",{"id":176,"version":177,"summary_zh":178,"released_at":179},89967,"v1.9","修复用于学习线性动态系统模型的EM算法，并改进相关功能。","2018-11-29T21:59:03",{"id":181,"version":182,"summary_zh":183,"released_at":184},89968,"v1.8","修复MRF、MLP神经网络及一些小问题。","2018-11-20T11:18:01",{"id":186,"version":187,"summary_zh":188,"released_at":189},89969,"v1.7","修复稀疏线性模型中的错误","2018-03-13T20:41:01",{"id":191,"version":192,"summary_zh":193,"released_at":194},89970,"v1.6","小幅改进","2017-12-28T05:43:01",{"id":196,"version":197,"summary_zh":198,"released_at":199},89971,"v1.5","修复 MLP 中的 minor bugs","2017-06-04T14:57:22",{"id":201,"version":202,"summary_zh":203,"released_at":204},89972,"v1.4","Added mean field, belief propagation and expectation propagation algorithms for Markov random field (MRF) model.","2017-05-28T07:19:57",{"id":206,"version":207,"summary_zh":208,"released_at":209},89973,"v1.3","some enhancement and minor bug fix","2017-04-19T17:43:21",{"id":211,"version":212,"summary_zh":213,"released_at":214},89974,"v1.2","Tweak some functions and fix some bugs.","2017-03-11T16:12:55",{"id":216,"version":217,"summary_zh":218,"released_at":219},89975,"v1.1","minor tweak and bug fix\r\n","2017-02-26T16:20:08",{"id":221,"version":222,"summary_zh":223,"released_at":224},89976,"v1.0","The file structure of the package is reorganised. I have done a lot of work to refactorize the code. I am satisfied with the quality of the code. The package is also thoroughly tested. All demos passed. Please now try, use and read it. Hope you enjoy my work. If you find any bug, please submit issue. I will do my best to fix when I have time.\r\n","2016-03-20T16:22:34",{"id":226,"version":227,"summary_zh":228,"released_at":229},89977,"v0.99","Last step before final release\n","2016-03-07T11:30:28",{"id":231,"version":232,"summary_zh":233,"released_at":234},89978,"v0.9","Currently, this package is in stable status. All functions has been refactorized and tested. Demos and doc for all function are provided and improved. Only things left are some feature, such as functions for NN, BP and EP. I encourage all people start to use this package from this release.\n","2016-02-21T04:05:00",{"id":236,"version":237,"summary_zh":238,"released_at":239},89979,"v0.8","Most of hard work has been done. Only documentation improvement and some coding style refinement are needed.\n","2016-02-18T12:21:24",{"id":241,"version":242,"summary_zh":243,"released_at":244},89980,"v0.5","Most of the algorithm functions and API are stable by now. Demos and auxilary functions are still needed for some chapters (mainly ch12, 13). After polishing chapter12, 13 and adding algorithms to chapter5, 8, this toolbox will considered finished.\r\n","2016-01-26T07:19:57",{"id":246,"version":247,"summary_zh":248,"released_at":249},89981,"v0.1","Initial release.\n","2015-10-22T11:18:39"]