[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"similar-autonomio--talos":3,"tool-autonomio--talos":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":76,"owner_avatar_url":77,"owner_bio":78,"owner_company":79,"owner_location":79,"owner_email":79,"owner_twitter":79,"owner_website":80,"owner_url":81,"languages":82,"stars":91,"forks":92,"last_commit_at":93,"license":94,"difficulty_score":23,"env_os":95,"env_gpu":96,"env_ram":96,"env_deps":97,"category_tags":103,"github_topics":104,"view_count":10,"oss_zip_url":79,"oss_zip_packed_at":79,"status":16,"created_at":109,"updated_at":110,"faqs":111,"releases":137},926,"autonomio\u002Ftalos","talos","Hyperparameter Experiments with TensorFlow and Keras","Talos 是一款专为 TensorFlow 和 Keras 设计的超参数自动优化工具，让神经网络调参变得简单高效。它能在不改动原有模型代码结构的前提下，自动完成超参数搜索、模型评估和结果分析，彻底解决手动\"炼丹\"调参的低效问题。\n\n研究人员和数据科学家是 Talos 的主要受众。如果你已经熟悉 Keras 或 TensorFlow，但厌倦了反复尝试学习率、批次大小、网络层数等参数的组合，Talos 可以让你几分钟内搭建完整的优化流程，无需学习新语法。它同样适合需要快速验证模型效果的工程师，以及希望系统探索参数空间的学术研究者。\n\nTalos 的核心亮点在于\"零侵入\"设计——直接包裹现有模型，不隐藏底层框架的任何功能。它支持网格搜索、随机搜索、概率优化等多种策略，甚至允许实验过程中动态切换优化方法。独特的\"人机协作\"模式让你可以随时介入调整方向。自 2019 年起稳定运行无重大缺陷，单条命令即可完成从优化到预测的完整流程，并自动生成实验分析报告。跨平台支持 CPU、GPU 及多卡环境。","\u003Ch1 align=\"center\">\r\n  \u003Cbr>\r\n  \u003Ca href=\"http:\u002F\u002Fautonom.io\">\u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fautonomio_talos_readme_651ec093bc56.png\" alt=\"Talos\" width=\"350\">\u003C\u002Fa>\r\n  \u003Cbr>\r\n\u003C\u002Fh1>\r\n\r\n\u003Ch3 align=\"center\">Bullet-Proof Hyperparameter Experiments with TensorFlow and Keras\u003C\u002Fh3>\r\n\r\n\u003Cp align=\"center\">\r\n  \u003Ca href=\"#talos\">Talos\u003C\u002Fa> •\r\n  \u003Ca href=\"#wrench-key-features\">Key Features\u003C\u002Fa> •\r\n  \u003Ca href=\"#arrow_forward-examples\">Examples\u003C\u002Fa> •\r\n  \u003Ca href=\"#floppy_disk-install\">Install\u003C\u002Fa> •\r\n  \u003Ca href=\"#speech_balloon-how-to-get-support\">Support\u003C\u002Fa> •\r\n  \u003Ca href=\"https:\u002F\u002Fautonomio.github.io\u002Ftalos\u002F\">Docs\u003C\u002Fa> •\r\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fautonomio\u002Ftalos\u002Fissues\">Issues\u003C\u002Fa> •\r\n  \u003Ca href=\"#page_with_curl-license\">License\u003C\u002Fa> •\r\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fautonomio\u002Ftalos\u002Farchive\u002Fmaster.zip\">Download\u003C\u002Fa>\r\n\u003C\u002Fp>\r\n\u003Chr>\r\n\u003Cp align=\"center\">\r\nTalos importantly improves ordinary TensorFlow (tf.keras) and Keras workflows  by \u003Cstrong>fully automating hyperparameter experiments\u003C\u002Fstrong> and \u003Cstrong>model evaluation\u003C\u002Fstrong>. Talos exposes TensorFlow (tf.keras) and Keras functionality entirely and there is no new syntax or templates to learn.\r\n\u003C\u002Fp>\r\n\u003Cp align=\"center\">\r\n\u003Cimg src='https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fautonomio_talos_readme_75fc5a82a558.gif' width=550px>\u003Cbr>\r\n\u003Cem>The above animation illustrates how a minimal Sequntial model is modified for Talos\u003C\u002Fem>\r\n\u003C\u002Fp>\r\n\r\n### Talos\r\n\r\nTL;DR Thousands of researchers have found Talos to importantly improve ordinary TensorFlow (tf.keras) and Keras workflows without taking away or hiding any of their power.\r\n\r\n  - Works with ANY Keras, TensorFlow (tf.keras) or PyTorch model\r\n  - Takes minutes to implement\r\n  - No new syntax to learn\r\n  - Adds zero new overhead to your workflow\r\n  - Bullet-proof results with no breaking bugs since 2019\r\n  - Comprehensive, up-to-date documentation\r\n\r\nTalos is made for researchers, data scientists, and data engineers that want to remain in **complete control of their TensorFlow (tf.keras) and Keras models**, but are tired of mindless parameter hopping and confusing optimization solutions that add complexity instead of reducing it. \r\n\r\n\u003Chr>\r\n\r\n### :wrench: Key Features\r\n\r\n**Within minutes, without learning any new syntax,** Talos allows you to configure, perform, and evaluate hyperparameter experiments that yield state-of-the-art results across a wide range of prediction tasks. Talos provides the **simplest and yet most powerful** available method for hyperparameter optimization with TensorFlow (tf.keras) and Keras. Key features include:\r\n\r\n  - Single-line optimize-to-predict pipeline `talos.Scan(x, y, model, params).predict(x_test, y_test)`\r\n  - Automated hyperparameter optimization\r\n  - Model generalization evaluator\r\n  - Experiment analytics\r\n  - Pseudo, Quasi, and Quantum Random search options\r\n  - Grid search\r\n  - Probabilistic optimizers\r\n  - Single file custom optimization strategies\r\n  - Dynamically change optimization strategy during experiment\r\n  - Support for man-machine cooperative optimization strategy\r\n  - Model candidate generality evaluation\r\n  - Live training monitor\r\n  - Experiment analytics\r\n\r\nTalos works on **Linux, Mac OSX**, and **Windows** systems and can be operated cpu, gpu, and multi-gpu systems.\r\n\r\n\u003Chr>\r\n\r\n### :arrow_forward: Examples\r\n\r\nGet the below code [here](https:\u002F\u002Fgist.github.com\u002Fmikkokotila\u002F4c0d6298ff0a22dc561fb387a1b4b0bb). More examples further below.\r\n\r\n\u003Cimg src=https:\u002F\u002Fi.ibb.co\u002FVWd8Bhm\u002FScreen-Shot-2019-01-06-at-11-26-32-PM.png>\r\n\r\nThe *Simple* example below is more than enough for starting to use Talos with any Keras model. *Field Report* has +4,400 claps on Medium because it's more entertaining.\r\n\r\n[Simple](https:\u002F\u002Fnbviewer.jupyter.org\u002Fgithub\u002Fautonomio\u002Ftalos\u002Fblob\u002Fmaster\u002Fexamples\u002FA%20Very%20Short%20Introduction%20to%20Hyperparameter%20Optimization%20of%20Keras%20Models%20with%20Talos.ipynb)  [1-2 mins]\r\n\r\n[Concise](https:\u002F\u002Fnbviewer.jupyter.org\u002Fgithub\u002Fautonomio\u002Ftalos\u002Fblob\u002Fmaster\u002Fexamples\u002FHyperparameter%20Optimization%20on%20Keras%20with%20Breast%20Cancer%20Data.ipynb)  [~5 mins]\r\n\r\n[Comprehensive](https:\u002F\u002Fnbviewer.jupyter.org\u002Fgithub\u002Fautonomio\u002Ftalos\u002Fblob\u002Fmaster\u002Fexamples\u002FHyperparameter%20Optimization%20with%20Keras%20for%20the%20Iris%20Prediction.ipynb)  [~10 mins]\r\n\r\n[Field Report](https:\u002F\u002Ftowardsdatascience.com\u002Fhyperparameter-optimization-with-keras-b82e6364ca53)  [~15 mins]\r\n\r\nFor more information on how Talos can help with your Keras, TensorFlow (tf.keras) and PyTorch workflow, visit the [User Manual](https:\u002F\u002Fautonomio.github.io\u002Ftalos\u002F).\r\n\r\nYou may also want to check out a visualization of the [Talos Hyperparameter Tuning workflow](https:\u002F\u002Fgithub.com\u002Fautonomio\u002Ftalos\u002Fwiki\u002FWorkflow).\r\n\r\n\u003Chr>\r\n\r\n### :floppy_disk: Install\r\n\r\nStable version:\r\n\r\n#### `pip install talos`\r\n\r\nDaily development version:\r\n\r\n#### `pip install git+https:\u002F\u002Fgithub.com\u002Fautonomio\u002Ftalos`\r\n\r\n\u003Chr>\r\n\r\n### :speech_balloon: How to get Support\r\n\r\n| I want to...                     | Go to...                                                  |\r\n| -------------------------------- | ---------------------------------------------------------- |\r\n| **...troubleshoot**           | [Docs] · [Wiki] · [GitHub Issue Tracker]                   |\r\n| **...report a bug**           | [GitHub Issue Tracker]                                     |\r\n| **...suggest a new feature**  | [GitHub Issue Tracker]                                     |\r\n| **...get support**            | [Stack Overflow]                     |\r\n\r\n\u003Chr>\r\n\r\n### :loudspeaker: Citations\r\n\r\nIf you use Talos for published work, please cite:\r\n\r\n`Autonomio Talos [Computer software]. (2024). Retrieved from http:\u002F\u002Fgithub.com\u002Fautonomio\u002Ftalos.`\r\n\r\n\u003Chr>\r\n\r\n### :page_with_curl: License\r\n\r\n[MIT License](https:\u002F\u002Fgithub.com\u002Fautonomio\u002Ftalos\u002Fblob\u002Fmaster\u002FLICENSE)\r\n\r\n[github issue tracker]: https:\u002F\u002Fgithub.com\u002Fautonomio\u002Ftalos\u002Fissues\r\n[docs]: https:\u002F\u002Fautonomio.github.io\u002Ftalos\u002F\r\n[wiki]: https:\u002F\u002Fgithub.com\u002Fautonomio\u002Ftalos\u002Fwiki\r\n[stack overflow]: https:\u002F\u002Fstackoverflow.com\u002Fquestions\u002Ftagged\u002Ftalos\r\n","\u003Ch1 align=\"center\">\n  \u003Cbr>\n  \u003Ca href=\"http:\u002F\u002Fautonom.io\">\u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fautonomio_talos_readme_651ec093bc56.png\" alt=\"Talos\" width=\"350\">\u003C\u002Fa>\n  \u003Cbr>\n\u003C\u002Fh1>\n\n\u003Ch3 align=\"center\">使用 TensorFlow 和 Keras 进行无懈可击的超参数实验\u003C\u002Fh3>\n\n\u003Cp align=\"center\">\n  \u003Ca href=\"#talos\">Talos\u003C\u002Fa> •\n  \u003Ca href=\"#wrench-主要功能\">主要功能\u003C\u002Fa> •\n  \u003Ca href=\"#arrow_forward-示例\">示例\u003C\u002Fa> •\n  \u003Ca href=\"#floppy_disk-安装\">安装\u003C\u002Fa> •\n  \u003Ca href=\"#speech_balloon-如何获取支持\">支持\u003C\u002Fa> •\n  \u003Ca href=\"https:\u002F\u002Fautonomio.github.io\u002Ftalos\u002F\">文档\u003C\u002Fa> •\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fautonomio\u002Ftalos\u002Fissues\">问题\u003C\u002Fa> •\n  \u003Ca href=\"#page_with_curl-许可证\">许可证\u003C\u002Fa> •\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fautonomio\u002Ftalos\u002Farchive\u002Fmaster.zip\">下载\u003C\u002Fa>\n\u003C\u002Fp>\n\u003Chr>\n\u003Cp align=\"center\">\nTalos 通过\u003Cstrong>完全自动化的超参数实验\u003C\u002Fstrong>和\u003Cstrong>模型评估\u003C\u002Fstrong>，显著改进了普通的 TensorFlow (tf.keras) 和 Keras 工作流程。Talos 完全暴露了 TensorFlow (tf.keras) 和 Keras 的功能，无需学习任何新语法或模板。\n\u003C\u002Fp>\n\u003Cp align=\"center\">\n\u003Cimg src='https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fautonomio_talos_readme_75fc5a82a558.gif' width=550px>\u003Cbr>\n\u003Cem>上面的动画展示了如何将一个最小的顺序（Sequential）模型修改为适用于 Talos 的形式\u003C\u002Fem>\n\u003C\u002Fp>\n\n### Talos\n\n简而言之，成千上万的研究人员发现 Talos 能够显著改进普通的 TensorFlow (tf.keras) 和 Keras 工作流程，同时不会削弱或隐藏它们的功能。\n\n  - 支持任何 Keras、TensorFlow (tf.keras) 或 PyTorch 模型\n  - 几分钟内即可实现\n  - 无需学习新语法\n  - 不会为您的工作流程增加额外负担\n  - 自 2019 年以来无重大错误，结果可靠\n  - 文档全面且保持最新\n\nTalos 专为研究人员、数据科学家和数据工程师设计，他们希望在**完全掌控自己的 TensorFlow (tf.keras) 和 Keras 模型**的同时，摆脱繁琐的参数调整和令人困惑的优化解决方案，这些方案往往增加了复杂性而非减少它。\n\n\u003Chr>\n\n### :wrench: 主要功能\n\n**几分钟内，无需学习任何新语法，** Talos 让您能够配置、执行和评估超参数实验，从而在广泛的预测任务中获得最先进的结果。Talos 提供了**最简单但最强大**的 TensorFlow (tf.keras) 和 Keras 超参数优化方法。主要功能包括：\n\n  - 单行优化到预测管道 `talos.Scan(x, y, model, params).predict(x_test, y_test)`\n  - 自动化超参数优化\n  - 模型泛化评估器\n  - 实验分析工具\n  - 伪随机、准随机和量子随机搜索选项\n  - 网格搜索\n  - 概率优化器\n  - 单文件自定义优化策略\n  - 在实验过程中动态更改优化策略\n  - 支持人机协作优化策略\n  - 模型候选泛化评估\n  - 实时训练监控\n  - 实验分析工具\n\nTalos 可在 **Linux、Mac OSX** 和 **Windows** 系统上运行，并支持 CPU、GPU 和多 GPU 系统。\n\n\u003Chr>\n\n### :arrow_forward: 示例\n\n点击 [这里](https:\u002F\u002Fgist.github.com\u002Fmikkokotila\u002F4c0d6298ff0a22dc561fb387a1b4b0bb) 获取以下代码。更多示例如下。\n\n\u003Cimg src=https:\u002F\u002Fi.ibb.co\u002FVWd8Bhm\u002FScreen-Shot-2019-01-06-at-11-26-32-PM.png>\n\n下面的 *Simple* 示例足以让您开始在任何 Keras 模型中使用 Talos。*Field Report* 在 Medium 上获得了超过 4,400 次点赞，因为它更有趣。\n\n[Simple](https:\u002F\u002Fnbviewer.jupyter.org\u002Fgithub\u002Fautonomio\u002Ftalos\u002Fblob\u002Fmaster\u002Fexamples\u002FA%20Very%20Short%20Introduction%20to%20Hyperparameter%20Optimization%20of%20Keras%20Models%20with%20Talos.ipynb)  [1-2 分钟]\n\n[Concise](https:\u002F\u002Fnbviewer.jupyter.org\u002Fgithub\u002Fautonomio\u002Ftalos\u002Fblob\u002Fmaster\u002Fexamples\u002FHyperparameter%20Optimization%20on%20Keras%20with%20Breast%20Cancer%20Data.ipynb)  [~5 分钟]\n\n[Comprehensive](https:\u002F\u002Fnbviewer.jupyter.org\u002Fgithub\u002Fautonomio\u002Ftalos\u002Fblob\u002Fmaster\u002Fexamples\u002FHyperparameter%20Optimization%20with%20Keras%20for%20the%20Iris%20Prediction.ipynb)  [~10 分钟]\n\n[Field Report](https:\u002F\u002Ftowardsdatascience.com\u002Fhyperparameter-optimization-with-keras-b82e6364ca53)  [~15 分钟]\n\n如需了解更多关于 Talos 如何帮助您的 Keras、TensorFlow (tf.keras) 和 PyTorch 工作流程的信息，请访问 [用户手册](https:\u002F\u002Fautonomio.github.io\u002Ftalos\u002F)。\n\n您还可以查看 [Talos 超参数调优工作流程](https:\u002F\u002Fgithub.com\u002Fautonomio\u002Ftalos\u002Fwiki\u002FWorkflow) 的可视化。\n\n\u003Chr>\n\n### :floppy_disk: 安装\n\n稳定版本：\n\n#### `pip install talos`\n\n每日开发版本：\n\n#### `pip install git+https:\u002F\u002Fgithub.com\u002Fautonomio\u002Ftalos`\n\n\u003Chr>\n\n### :speech_balloon: 如何获取支持\n\n| 我想...                     | 去...                                                  |\n| -------------------------------- | ---------------------------------------------------------- |\n| **...排查问题**           | [文档] · [Wiki] · [GitHub 问题跟踪器]                   |\n| **...报告错误**           | [GitHub 问题跟踪器]                                     |\n| **...建议新功能**         | [GitHub 问题跟踪器]                                     |\n| **...获取支持**           | [Stack Overflow]                                         |\n\n\u003Chr>\n\n### :loudspeaker: 引用\n\n如果您在已发表的作品中使用了 Talos，请引用：\n\n`Autonomio Talos [计算机软件]. (2024). 取自 http:\u002F\u002Fgithub.com\u002Fautonomio\u002Ftalos.`\n\n\u003Chr>\n\n### :page_with_curl: 许可证\n\n[MIT 许可证](https:\u002F\u002Fgithub.com\u002Fautonomio\u002Ftalos\u002Fblob\u002Fmaster\u002FLICENSE)\n\n[github issue tracker]: https:\u002F\u002Fgithub.com\u002Fautonomio\u002Ftalos\u002Fissues\n[docs]: https:\u002F\u002Fautonomio.github.io\u002Ftalos\u002F\n[wiki]: https:\u002F\u002Fgithub.com\u002Fautonomio\u002Ftalos\u002Fwiki\n[stack overflow]: https:\u002F\u002Fstackoverflow.com\u002Fquestions\u002Ftagged\u002Ftalos","# Talos 快速上手指南\n\nTalos 是一个强大的工具，用于自动化 TensorFlow (tf.keras) 和 Keras 的超参数优化和模型评估。它无需学习新语法，完全兼容现有的工作流。\n\n---\n\n## 环境准备\n\n### 系统要求\n- 支持的操作系统：Linux、Mac OSX、Windows\n- 支持的硬件：CPU、GPU、多 GPU 系统\n\n### 前置依赖\n- Python 3.6 或更高版本\n- 已安装 TensorFlow 或 Keras 模型环境\n\n---\n\n## 安装步骤\n\n### 稳定版安装\n运行以下命令以安装最新稳定版本：\n```bash\npip install talos\n```\n\n### 开发版安装\n如果需要使用最新的开发版本，可以通过以下命令安装：\n```bash\npip install git+https:\u002F\u002Fgithub.com\u002Fautonomio\u002Ftalos\n```\n\n> **提示**：如果在中国大陆地区，建议使用国内镜像源加速安装，例如：\n> ```bash\n> pip install talos -i https:\u002F\u002Fpypi.tuna.tsinghua.edu.cn\u002Fsimple\n> ```\n\n---\n\n## 基本使用\n\n以下是一个最简单的使用示例，展示如何将 Talos 集成到现有的 Keras 模型中。\n\n### 示例代码\n```python\nimport talos\nfrom tensorflow.keras.models import Sequential\nfrom tensorflow.keras.layers import Dense\n\n# 定义模型\ndef create_model(x_train, y_train, x_val, y_val, params):\n    model = Sequential()\n    model.add(Dense(params['units'], input_dim=x_train.shape[1], activation='relu'))\n    model.add(Dense(1, activation='sigmoid'))\n    model.compile(optimizer=params['optimizer'], loss='binary_crossentropy', metrics=['accuracy'])\n    history = model.fit(x_train, y_train, validation_data=(x_val, y_val), epochs=params['epochs'], batch_size=params['batch_size'], verbose=0)\n    return history, model\n\n# 定义超参数搜索空间\np = {\n    'units': [16, 32, 64],\n    'optimizer': ['adam', 'sgd'],\n    'epochs': [10, 20],\n    'batch_size': [32, 64]\n}\n\n# 执行超参数优化\nscan_object = talos.Scan(x=x_train, y=y_train, model=create_model, params=p, experiment_name='talos_example')\n\n# 使用优化结果进行预测\npredict_object = scan_object.predict(x_test, y_test)\n```\n\n### 运行说明\n1. 将 `x_train`, `y_train`, `x_test`, `y_test` 替换为实际的数据集。\n2. 根据需求调整超参数搜索空间 `p`。\n3. 运行脚本后，Talos 会自动执行超参数优化并输出最佳模型。\n\n---\n\n以上是 Talos 的快速上手指南，更多高级功能和详细文档请参考 [官方文档](https:\u002F\u002Fautonomio.github.io\u002Ftalos\u002F)。","一位数据科学家正在为一家电商公司开发一个基于 Keras 的商品推荐模型，希望通过优化超参数提升模型性能。\n\n### 没有 talos 时\n- 手动调整超参数非常耗时，每次修改后都需要重新运行整个训练流程，效率低下。\n- 难以系统性地记录和比较不同超参数组合的实验结果，容易遗漏最佳配置。\n- 对模型的泛化能力评估不够全面，经常出现训练效果好但测试效果差的情况。\n- 缺乏直观的实验监控工具，无法实时了解训练过程中的关键指标变化。\n- 尝试不同的优化策略需要编写大量额外代码，增加了开发复杂度。\n\n### 使用 talos 后\n- 只需添加一行代码即可实现自动化超参数优化，大幅缩短了实验周期。\n- 自动生成详细的实验报告，方便对比和分析不同参数组合的效果，确保找到最优解。\n- 内置模型泛化能力评估功能，帮助快速筛选出表现稳定的模型候选。\n- 提供实时训练监控界面，可以随时查看损失值、准确率等关键指标的变化趋势。\n- 支持多种优化策略且可动态切换，无需额外编码即可灵活尝试不同方法。\n\n通过 talos，数据科学家能够专注于模型设计本身，而将繁琐的超参数调优工作交给工具自动完成，显著提升了开发效率和模型质量。","https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fautonomio_talos_651ec093.png","autonomio","Autonomio","https:\u002F\u002Foss.gittoolsai.com\u002Favatars\u002Fautonomio_047db17a.png","Machine Intelligence Workbench",null,"http:\u002F\u002Fautonom.io","https:\u002F\u002Fgithub.com\u002Fautonomio",[83,87],{"name":84,"color":85,"percentage":86},"Python","#3572A5",99.8,{"name":88,"color":89,"percentage":90},"Shell","#89e051",0.2,1637,266,"2026-04-02T08:37:14","MIT","Linux, macOS, Windows","未说明",{"notes":98,"python":96,"dependencies":99},"支持 CPU、GPU 和多 GPU 系统，具体需求视模型规模而定；建议参考官方文档以获取更详细的依赖信息。",[100,101,102],"tensorflow","keras","pytorch",[13],[105,101,106,107,108,100],"deep-learning","keras-tensorflow","hyperparameter-optimization","artificial-intelligence","2026-03-27T02:49:30.150509","2026-04-06T07:11:58.461202",[112,117,122,127,132],{"id":113,"question_zh":114,"answer_zh":115,"source_url":116},4055,"如何解决 Talos 使用 TensorFlow 后端时的内存泄漏问题？","可以尝试在每次迭代后调用 `tf.reset_default_graph()` 来释放未使用的计算图资源。此外，确保启用了 `clear_tf_session=True` 参数，并结合 `gc.collect()` 手动触发垃圾回收。","https:\u002F\u002Fgithub.com\u002Fautonomio\u002Ftalos\u002Fissues\u002F343",{"id":118,"question_zh":119,"answer_zh":120,"source_url":121},4051,"如何在 Talos 中使用生成器 (generator) 进行超参数优化？","可以通过 Johan 提供的示例代码实现，他在博客中详细介绍了如何将 Keras 的生成器与 Talos 结合使用。虽然不是最新版本的 Talos，但可以轻松适配。参考链接：[Johan 的博客](http:\u002F\u002Finquisitive.blog\u002Fpost\u002F2019-08-08-11-59-Hyperparameter-optimization-with-Keras-generators-and-Talos)。","https:\u002F\u002Fgithub.com\u002Fautonomio\u002Ftalos\u002Fissues\u002F11",{"id":123,"question_zh":124,"answer_zh":125,"source_url":126},4052,"如何解决 Talos 在多次迭代后出现 ResourceExhaustedError 的问题？","可以通过禁用 TensorFlow 的 eager execution 模式来解决此问题。具体方法是在代码中添加以下命令：`tf.compat.v1.disable_eager_execution()`。","https:\u002F\u002Fgithub.com\u002Fautonomio\u002Ftalos\u002Fissues\u002F482",{"id":128,"question_zh":129,"answer_zh":130,"source_url":131},4053,"如何在 Talos 中传递多个输入（如列表形式）给模型？","目前 Talos 不直接支持传递列表形式的输入，但可以通过设置 `x_val` 和 `y_val` 参数避免验证集拆分的问题。如果需要交叉验证，建议查看相关 Issue（如 #155、#145、#178）以获取更多信息。","https:\u002F\u002Fgithub.com\u002Fautonomio\u002Ftalos\u002Fissues\u002F132",{"id":133,"question_zh":134,"answer_zh":135,"source_url":136},4054,"为什么 Talos 报告中的列顺序不正确？","这是由于 Python 2.7 中字典的无序性导致的。维护者计划通过使用 `OrderedDict` 来修复此问题。如果您遇到类似问题，可以尝试升级到更高版本的 Talos 或切换到 Python 3.x。","https:\u002F\u002Fgithub.com\u002Fautonomio\u002Ftalos\u002Fissues\u002F146",[138,143,148,153,158,163,168,173,178,183,187,192,197,201,205,209],{"id":139,"version":140,"summary_zh":141,"released_at":142},103477,"v1.4","## What changes?\r\n\r\n- Migrates package management to Hatch \r\n- Migrates Deploy CI to Hatch\r\n- Cleanup varios compatibility issues with later versions of Tensorflow\r\n- Lock Tensorflow to `2.14.1` \r\n\r\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002Fautonomio\u002Ftalos\u002Fcompare\u002Fv1.3...v1.4","2024-04-21T09:02:06",{"id":144,"version":145,"summary_zh":146,"released_at":147},103478,"v1.3","## Major Changes\r\n\r\n- Added comprehensive support for multi-input model workflows\r\n- Added `DistributeParamSpace` to allow distributing parameter space across several hosts\r\n\r\n## Fixes\r\n\r\n- Fixed an issue that would mix the order of columns on the experiment log in Python 3.6 and older\r\n\r\n## Minor Changes\r\n\r\n- Added more control over the \r\n- Unified naming convention of all Callbacks\r\n- Moved all Callbacks to `callbacks` submodule\r\n","2022-05-28T10:07:13",{"id":149,"version":150,"summary_zh":151,"released_at":152},103479,"v1.2.3","- Unifies the way parameter space is created\r\n- Fixes issues with PowerDrawCallback\r\n- Adds DistributeParamSpace for distributed hyperparameter experiments\r\n- Several other small fixes and cleanups","2022-04-15T14:37:02",{"id":154,"version":155,"summary_zh":156,"released_at":157},103480,"v1.0.2","Small fixes to bring everything up-to-date.","2022-01-28T21:35:24",{"id":159,"version":160,"summary_zh":161,"released_at":162},103481,"v1.0","The initial release of the new multi-backend Talos.","2020-11-09T16:48:30",{"id":164,"version":165,"summary_zh":166,"released_at":167},103482,"v0.6.7","Supports multi-backend Keras.","2020-11-09T16:46:05",{"id":169,"version":170,"summary_zh":171,"released_at":172},103483,"v0.6.6","Fixes a version compatibility issue with TensorFlow.","2020-01-25T19:32:41",{"id":174,"version":175,"summary_zh":176,"released_at":177},103484,"v0.6.4","A large number of fixes to `v.0.6.3`. Moving to semantic versioning in this release.","2020-01-25T18:58:46",{"id":179,"version":180,"summary_zh":181,"released_at":182},103485,"v.0.6.3","This release starts a new LTS and a new era in Talos story.","2019-08-16T19:51:33",{"id":184,"version":185,"summary_zh":79,"released_at":186},103486,"v.0.5","2019-05-27T05:52:15",{"id":188,"version":189,"summary_zh":190,"released_at":191},103487,"v.0.4.9","This version fixes several important issues and introduces many new features.\r\n\r\n","2019-03-02T10:03:37",{"id":193,"version":194,"summary_zh":195,"released_at":196},103488,"v.0.4.8","This version fixes several important issues and introduces many new features.","2019-02-22T16:05:01",{"id":198,"version":199,"summary_zh":195,"released_at":200},103489,"v.0.4.7","2019-02-21T09:26:17",{"id":202,"version":203,"summary_zh":79,"released_at":204},103490,"v.0.4.4","2018-10-05T20:32:20",{"id":206,"version":207,"summary_zh":79,"released_at":208},103491,"v.0.1.9.5","2018-07-28T12:17:32",{"id":210,"version":211,"summary_zh":79,"released_at":212},103492,"v0.1.8","2018-05-11T16:07:00"]