[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"similar-fastai--course20":3,"tool-fastai--course20":61},[4,18,26,36,44,53],{"id":5,"name":6,"github_repo":7,"description_zh":8,"stars":9,"difficulty_score":10,"last_commit_at":11,"category_tags":12,"status":17},4358,"openclaw","openclaw\u002Fopenclaw","OpenClaw 是一款专为个人打造的本地化 AI 助手，旨在让你在自己的设备上拥有完全可控的智能伙伴。它打破了传统 AI 助手局限于特定网页或应用的束缚，能够直接接入你日常使用的各类通讯渠道，包括微信、WhatsApp、Telegram、Discord、iMessage 等数十种平台。无论你在哪个聊天软件中发送消息，OpenClaw 都能即时响应，甚至支持在 macOS、iOS 和 Android 设备上进行语音交互，并提供实时的画布渲染功能供你操控。\n\n这款工具主要解决了用户对数据隐私、响应速度以及“始终在线”体验的需求。通过将 AI 部署在本地，用户无需依赖云端服务即可享受快速、私密的智能辅助，真正实现了“你的数据，你做主”。其独特的技术亮点在于强大的网关架构，将控制平面与核心助手分离，确保跨平台通信的流畅性与扩展性。\n\nOpenClaw 非常适合希望构建个性化工作流的技术爱好者、开发者，以及注重隐私保护且不愿被单一生态绑定的普通用户。只要具备基础的终端操作能力（支持 macOS、Linux 及 Windows WSL2），即可通过简单的命令行引导完成部署。如果你渴望拥有一个懂你",349277,3,"2026-04-06T06:32:30",[13,14,15,16],"Agent","开发框架","图像","数据工具","ready",{"id":19,"name":20,"github_repo":21,"description_zh":22,"stars":23,"difficulty_score":10,"last_commit_at":24,"category_tags":25,"status":17},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,"2026-04-05T11:01:52",[14,15,13],{"id":27,"name":28,"github_repo":29,"description_zh":30,"stars":31,"difficulty_score":32,"last_commit_at":33,"category_tags":34,"status":17},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",155373,2,"2026-04-14T11:34:08",[14,13,35],"语言模型",{"id":37,"name":38,"github_repo":39,"description_zh":40,"stars":41,"difficulty_score":32,"last_commit_at":42,"category_tags":43,"status":17},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。",108322,"2026-04-10T11:39:34",[14,15,13],{"id":45,"name":46,"github_repo":47,"description_zh":48,"stars":49,"difficulty_score":32,"last_commit_at":50,"category_tags":51,"status":17},6121,"gemini-cli","google-gemini\u002Fgemini-cli","gemini-cli 是一款由谷歌推出的开源 AI 命令行工具，它将强大的 Gemini 大模型能力直接集成到用户的终端环境中。对于习惯在命令行工作的开发者而言，它提供了一条从输入提示词到获取模型响应的最短路径，无需切换窗口即可享受智能辅助。\n\n这款工具主要解决了开发过程中频繁上下文切换的痛点，让用户能在熟悉的终端界面内直接完成代码理解、生成、调试以及自动化运维任务。无论是查询大型代码库、根据草图生成应用，还是执行复杂的 Git 操作，gemini-cli 都能通过自然语言指令高效处理。\n\n它特别适合广大软件工程师、DevOps 人员及技术研究人员使用。其核心亮点包括支持高达 100 万 token 的超长上下文窗口，具备出色的逻辑推理能力；内置 Google 搜索、文件操作及 Shell 命令执行等实用工具；更独特的是，它支持 MCP（模型上下文协议），允许用户灵活扩展自定义集成，连接如图像生成等外部能力。此外，个人谷歌账号即可享受免费的额度支持，且项目基于 Apache 2.0 协议完全开源，是提升终端工作效率的理想助手。",100752,"2026-04-10T01:20:03",[52,13,15,14],"插件",{"id":54,"name":55,"github_repo":56,"description_zh":57,"stars":58,"difficulty_score":32,"last_commit_at":59,"category_tags":60,"status":17},4721,"markitdown","microsoft\u002Fmarkitdown","MarkItDown 是一款由微软 AutoGen 团队打造的轻量级 Python 工具，专为将各类文件高效转换为 Markdown 格式而设计。它支持 PDF、Word、Excel、PPT、图片（含 OCR）、音频（含语音转录）、HTML 乃至 YouTube 链接等多种格式的解析，能够精准提取文档中的标题、列表、表格和链接等关键结构信息。\n\n在人工智能应用日益普及的今天，大语言模型（LLM）虽擅长处理文本，却难以直接读取复杂的二进制办公文档。MarkItDown 恰好解决了这一痛点，它将非结构化或半结构化的文件转化为模型“原生理解”且 Token 效率极高的 Markdown 格式，成为连接本地文件与 AI 分析 pipeline 的理想桥梁。此外，它还提供了 MCP（模型上下文协议）服务器，可无缝集成到 Claude Desktop 等 LLM 应用中。\n\n这款工具特别适合开发者、数据科学家及 AI 研究人员使用，尤其是那些需要构建文档检索增强生成（RAG）系统、进行批量文本分析或希望让 AI 助手直接“阅读”本地文件的用户。虽然生成的内容也具备一定可读性，但其核心优势在于为机器",93400,"2026-04-06T19:52:38",[52,14],{"id":62,"github_repo":63,"name":64,"description_en":65,"description_zh":66,"ai_summary_zh":66,"readme_en":67,"readme_zh":68,"quickstart_zh":69,"use_case_zh":70,"hero_image_url":71,"owner_login":72,"owner_name":73,"owner_avatar_url":74,"owner_bio":75,"owner_company":76,"owner_location":76,"owner_email":76,"owner_twitter":77,"owner_website":78,"owner_url":79,"languages":80,"stars":109,"forks":110,"last_commit_at":111,"license":112,"difficulty_score":32,"env_os":113,"env_gpu":114,"env_ram":115,"env_deps":116,"category_tags":122,"github_topics":123,"view_count":32,"oss_zip_url":76,"oss_zip_packed_at":76,"status":17,"created_at":129,"updated_at":130,"faqs":131,"releases":167},7570,"fastai\u002Fcourse20","course20","Deep Learning for Coders, 2020, the website","course20 是一套专为程序员设计的深度学习实战课程，旨在让开发者无需博士学位也能掌握人工智能应用开发。它通过结合《Deep Learning for Coders》一书与 2020 版视频教程，解决了传统深度学习学习曲线陡峭、理论过于抽象的痛点，帮助用户快速从代码实践入手构建模型。\n\n这套资源非常适合具备一定编程基础但缺乏机器学习背景的开发者，同时也适合希望快速上手解决实际问题的数据科学家。课程独特的亮点在于“自顶向下”的教学法：先让用户运行并修改现成的强大模型以获得直观反馈，再逐步深入讲解背后的数学原理。内容基于 fastai 库和 PyTorch 框架，所有章节均提供可交互的 Jupyter Notebook 代码实例，支持直接在云端环境（如 Gradient 或 Colab）中运行实验，无需复杂的环境配置。此外，视频提供多语言字幕（含简体中文）及全文检索功能，配合活跃的全球社区论坛，为学习者提供了友好的互助环境。无论你是想转行 AI 的工程师，还是希望将智能功能融入产品的技术人员，course20 都能为你提供一条高效、务实的学习路径。","# Practical Deep Learning for Coders\n> Deep Learning for Coders with fastai and PyTorch: AI Applications Without a PhD - the book and the course\n\n\nWelcome to *Practical Deep Learning for Coders*. This web site covers the book and the 2020 version of the course, which are designed to work closely together. If you haven't yet got the book, you can [buy it here](https:\u002F\u002Fwww.amazon.com\u002FDeep-Learning-Coders-fastai-PyTorch\u002Fdp\u002F1492045527). It's also [freely available](https:\u002F\u002Fgithub.com\u002Ffastai\u002Ffastbook) as interactive Jupyter Notebooks; read on to learn how to access them..\n\n## How do I get started?\n\nIf you're ready to dive in right now, here's how to get started. If you want to know more about this course, read the next sections, and then come back here.\n\nTo watch the videos, click on the *Lessons* section in the navigation sidebar. The lessons all have searchable transcripts; click \"Transcript Search\" in the top right panel to search for a word or phrase, and then click it to jump straight to video at the time that appears in the transcript. The videos are all captioned and also translated into Chinese (简体中文) and Spanish; while watching the video click the \"CC\" button to turn them on and off, and the setting button to change the language.\n\nEach video covers a chapter from the book. The entirety of every chapter of the book is available as an interactive Jupyter Notebook. [Jupyter Notebook](https:\u002F\u002Fjupyter.org\u002F) is the most popular tool for doing data science in Python, for good reason. It is powerful, flexible, and easy to use. We think you will love it! Since the most important thing for learning deep learning is writing code and experimenting, it's important that you have a great platform for experimenting with code.\n\nTo get started, we recommend using a Jupyter Server from one of the recommended online platforms (click the links for instructions on how to use these for the course):\n- [Gradient](\u002Fstart_gradient): Unlike Colab, this is a \"real\" Jupyter Notebook so everything in the course works. It also provides space to save your notebooks and models. However, sometimes the free servers get over-loaded, and when that happens it's impossible to connect\n- [Colab](\u002Fstart_colab): A popular free service from Google. Google have changed the notebook platform quite a lot, so keyboard shortcuts are different, and not everything works (e.g. much of chapter 2 doesn't work because Colab doesn't support creating web app GUIs).\n\nIf you are interested in the experience of running a full Linux server you can consider [Google Cloud](start_gcp) (extremely popular service, very reliable, but the fastest GPUs are far more expensive). We strongly suggest using one of the recommended online platforms for running the notebooks, and to *not* use your own computer, unless you're very experienced with Linux system adminstration and handling GPU drivers, CUDA, and so forth.\n\nIf you need help, there's a [wonderful online community](https:\u002F\u002Fforums.fast.ai\u002Fc\u002Fpart1-v4\u002F46) ready to help you at forums.fast.ai. Before asking a question on the forums, search carefully to see if your question has been answered before. (The forum system won't let you post until you've spent a few minutes on the site reading existing topics.) One bit that many students find tricky is getting signed up for the Bing API for the image download task in lesson 2; here's a helpful [forum post](https:\u002F\u002Fforums.fast.ai\u002Ft\u002Fgetting-the-bing-image-search-key\u002F67417) explaining how to get the Bing API key you'll need for downloading images.\n\n## Is this course for me?\n\nThank you for letting us join you on your deep learning journey, however far along that you may be! Previous fast.ai courses have been studied by hundreds of thousands of students, from all walks of life, from all parts of the world. Many students have told us about how they've become [multiple gold medal winners](https:\u002F\u002Fforums.fast.ai\u002Ft\u002Fmy-first-gold-medal\u002F54237) of [international machine learning competitions](https:\u002F\u002Ftowardsdatascience.com\u002Fmy-3-year-journey-from-zero-python-to-deep-learning-competition-master-6605c188eec7), [received offers](https:\u002F\u002Fforums.fast.ai\u002Ft\u002Fhow-has-your-journey-been-so-far-learners\u002F6480\u002F2) from top companies, and having [research](https:\u002F\u002Fui.adsabs.harvard.edu\u002Fabs\u002F2020EGUGA..2221465A\u002Fabstract) [papers](http:\u002F\u002Fwww.ieomsociety.org\u002Fieom2020\u002Fpapers\u002F37.pdf) [published](https:\u002F\u002Fpubs.rsna.org\u002Fdoi\u002Fabs\u002F10.1148\u002Fryai.2019190113?journalCode=ai). For instance, Isaac Dimitrovsky [told us](https:\u002F\u002Fforums.fast.ai\u002Ft\u002Fthanks-ra2-dream-challenge-win\u002F76875) that he had \"*been playing around with ML for a couple of years without really grokking it... [then] went through the fast.ai part 1 course late last year, and it clicked for me*\". He went on to achieve first place in the prestigious international [RA2-DREAM Challenge](https:\u002F\u002Fwww.synapse.org\u002F#!Synapse:syn20545111\u002Fwiki\u002F594083) competition! He developed a [multistage deep learning method](https:\u002F\u002Fwww.synapse.org\u002F#!Synapse:syn21478998\u002Fwiki\u002F604432) for scoring radiographic hand and foot joint damage in rheumatoid arthritis, taking advantage of the fastai library.\n\nIt doesn't matter if you don't come from a technical or a mathematical background (though it's okay if you do too!); we wrote this course to make deep learning accessible to as many people as possible. The only prerequisite is that you know how to code (a year of experience is enough), preferably in Python, and that you have at least followed a high school math course. The first three chapters have been explicitly written in a way that will allow executives, product managers, etc. to understand the most important things they'll need to know about deep learning -- if that's you, just skip over the code in those sections.\n\nDeep learning is a computer technique to extract and transform data–-with use cases ranging from human speech recognition to animal imagery classification–-by using multiple layers of neural networks. A lot of people assume that you need all kinds of hard-to-find stuff to get great results with deep learning, but as you'll see in this course, those people are wrong. Here's a few things you *absolutely don't need* to do world-class deep learning:\n\n| Myth (don't need) | Truth\n|---|---|\n| Lots of math | Just high school math is sufficient\n| Lots of data | We've seen record-breaking results with \u003C50 items of data\n| Lots of expensive computers | You can get what you need for state of the art work for free\n\nDeep learning has power, flexibility, and simplicity. That's why we believe it should be applied across many disciplines. These include the social and physical sciences, the arts, medicine, finance, scientific research, and many more. Here's a list of some of the thousands of tasks in different areas at which deep learning, or methods heavily using deep learning, is now the best in the world:\n\n- **Natural language processing (NLP)** Answering questions; speech recognition; summarizing documents; classifying documents; finding names, dates, etc. in documents; searching for articles mentioning a concept\n- **Computer vision** Satellite and drone imagery interpretation (e.g., for disaster resilience); face recognition; image captioning; reading traffic signs; locating pedestrians and vehicles in autonomous vehicles\n- **Medicine** Finding anomalies in radiology images, including CT, MRI, and X-ray images; counting features in pathology slides; measuring features in ultrasounds; diagnosing diabetic retinopathy\n- **Biology** Folding proteins; classifying proteins; many genomics tasks, such as tumor-normal sequencing and classifying clinically actionable genetic mutations; cell classification; analyzing protein\u002Fprotein interactions\n- **Image generation** Colorizing images; increasing image resolution; removing noise from images; converting images to art in the style of famous artists\n- **Recommendation systems** Web search; product recommendations; home page layout\n- **Playing games** Chess, Go, most Atari video games, and many real-time strategy games\n- **Robotics** Handling objects that are challenging to locate (e.g., transparent, shiny, lacking texture) or hard to pick up\n- **Other applications** Financial and logistical forecasting, text to speech, and much more...\n\n## Who we are\n\nWe are Sylvain Gugger and Jeremy Howard, your guides on this journey. We're the co-authors of fastai, the software that you'll be using throughout this course.\n\nJeremy has been using and teaching machine learning for around 30 years. He started using neural networks 25 years ago. During this time, he has led many companies and projects that have machine learning at their core, including founding the first company to focus on deep learning and medicine, Enlitic, and taking on the role of President and Chief Scientist of the world's largest machine learning community, Kaggle. He is the co-founder, along with Dr. Rachel Thomas, of fast.ai, the organization that built the course this course is based on.\n\nSylvain has written 10 math textbooks, covering the entire advanced French maths curriculum! He is now a researcher at Hugging Face, and was previously a researcher at fast.ai.\n\nWe care a lot about teaching. In this course, we start by showing how to use a complete, working, very usable, state-of-the-art deep learning network to solve real-world problems, using simple, expressive tools. And then we gradually dig deeper and deeper into understanding how those tools are made, and how the tools that make those tools are made, and so on… We always teaching through examples. We ensure that there is a context and a purpose that you can understand intuitively, rather than starting with algebraic symbol manipulation.\n\n## The software you will be using\n\nIn this course, you'll be using [PyTorch](https:\u002F\u002Fpytorch.org\u002F) and [fastai](https:\u002F\u002Fdocs.fast.ai).\n\nWe've completed hundreds of machine learning projects using dozens of different packages, and many different programming languages. At fast.ai, we have written courses using most of the main deep learning and machine learning packages used today. We spent over a thousand hours testing PyTorch before deciding that we would use it for future courses, software development, and research. PyTorch is now the world's fastest-growing deep learning library and is already used for most research papers at top conferences.\n\nPyTorch works best as a low-level foundation library, providing the basic operations for higher-level functionality. The fastai library is the most popular library for adding this higher-level functionality on top of PyTorch. In this course, as we go deeper and deeper into the foundations of deep learning, we will also go deeper and deeper into the layers of fastai. This course covers version 2 of the fastai library, which is a from-scratch rewrite providing many unique features.\n\n## What you will learn\n\nAfter finishing this course you will know:\n\n- How to train models that achieve state-of-the-art results in:\n  - Computer vision, including image classification (e.g., classifying pet photos by breed), and image localization and detection (e.g., finding where the animals in an image are)\n  - Natural language processing (NLP), including document classification (e.g., movie review sentiment analysis) and language modeling\n  - Tabular data (e.g., sales prediction) with categorical data, continuous data, and mixed data, including time series\n  - Collaborative filtering (e.g., movie recommendation)\n- How to turn your models into web applications, and deploy them\n- Why and how deep learning models work, and how to use that knowledge to improve the accuracy, speed, and reliability of your models\n- The latest deep learning techniques that really matter in practice\n- How to implement stochastic gradient descent and a complete training loop from scratch\n- How to think about the ethical implications of your work, to help ensure that you're making the world a better place and that your work isn't misused for harm\n\nHere are some of the techniques covered (don't worry if none of these words mean anything to you yet--you'll learn them all soon): \n\n- Random forests and gradient boosting\n- Affine functions and nonlinearities\n- Parameters and activations\n- Random initialization and transfer learning\n- SGD, Momentum, Adam, and other optimizers\n- Convolutions\n- Batch normalization\n- Dropout\n- Data augmentation\n- Weight decay\n- Image classification and regression\n- Entity and word embeddings\n- Recurrent neural networks (RNNs)\n- Segmentation\n- And much more\n","# 针对编码者的实用深度学习\n> 使用 fastai 和 PyTorch 的编码者深度学习：无需博士学位的 AI 应用——本书及课程\n\n\n欢迎来到《针对编码者的实用深度学习》。本网站涵盖了本书以及 2020 年版课程的内容，两者设计为紧密配合使用。如果您尚未拥有本书，可以在此 [购买](https:\u002F\u002Fwww.amazon.com\u002FDeep-Learning-Coders-fastai-PyTorch\u002Fdp\u002F1492045527)。此外，本书也以交互式 Jupyter Notebook 的形式 [免费提供](https:\u002F\u002Fgithub.com\u002Ffastai\u002Ffastbook)；请继续阅读以了解如何获取它们。\n\n## 我该如何开始？\n\n如果您现在就准备开始学习，以下是入门步骤。如果您想了解更多关于本课程的信息，请先阅读接下来的几部分，然后再回到这里。\n\n要观看视频，请点击导航侧边栏中的“课程”部分。所有课程都配有可搜索的字幕文本；点击右上角的“字幕搜索”即可查找特定词汇或短语，然后单击该词句即可直接跳转到视频中出现该内容的时间点。视频均带有字幕，并已翻译成简体中文和西班牙语。观看时，您可以点击“CC”按钮来开启或关闭字幕，再点击设置按钮选择语言。\n\n每段视频对应书中的一章内容。而书中的每一章内容也都以交互式 Jupyter Notebook 的形式提供。[Jupyter Notebook](https:\u002F\u002Fjupyter.org\u002F) 是 Python 数据科学领域最受欢迎的工具，这绝非偶然——它功能强大、灵活且易于使用。我们相信您会爱上它！由于学习深度学习最重要的就是编写代码并进行实验，因此拥有一款优秀的代码实验平台至关重要。\n\n为了开始学习，我们推荐您使用以下推荐在线平台提供的 Jupyter 服务器（点击链接可查看如何在这些平台上运行本课程的相关说明）：\n- [Gradient](\u002Fstart_gradient)：与 Colab 不同，这是一个“真正的”Jupyter Notebook 环境，因此课程中的所有内容都能正常运行。它还提供存储笔记本和模型的空间。不过，有时免费服务器可能会过载，导致无法连接。\n- [Colab](\u002Fstart_colab)：谷歌推出的热门免费服务。由于谷歌近期对 Notebook 平台进行了较大改动，快捷键有所变化，且并非所有功能都可用（例如，第 2 章的许多内容无法运行，因为 Colab 不支持创建 Web 应用 GUI）。\n\n如果您希望体验完整 Linux 服务器的运行环境，也可以考虑 [Google Cloud](start_gcp)（非常流行且可靠的服务，但性能最强的 GPU 成本较高）。我们强烈建议您使用上述推荐的在线平台来运行笔记本，而不要在自己的电脑上操作，除非您具备丰富的 Linux 系统管理经验，并熟悉 GPU 驱动程序、CUDA 等相关技术。\n\n如果您需要帮助，forums.fast.ai 上有一个 [优秀的在线社区](https:\u002F\u002Fforums.fast.ai\u002Fc\u002Fpart1-v4\u002F46)，随时准备为您提供支持。在论坛发帖提问之前，请务必仔细搜索，看看类似问题是否已经有人回答过。（论坛系统会要求您先在网站上花几分钟阅读现有主题，才能发布新帖子。）许多学员觉得比较棘手的一个环节是为第 2 课中的图片下载任务注册必应 API 密钥；这里有一篇很有帮助的 [论坛帖子](https:\u002F\u002Fforums.fast.ai\u002Ft\u002Fgetting-the-bing-image-search-key\u002F67417)，详细介绍了如何获取下载图片所需的必应 API 密钥。\n\n## 这门课程适合我吗？\n\n感谢您邀请我们陪伴您踏上深度学习之旅，无论您目前处于哪个阶段！以往的 fast.ai 课程吸引了来自世界各地、各行各业的数十万名学员。许多学员分享了他们的成长经历：有人在国际机器学习竞赛中多次斩获金牌（[详情](https:\u002F\u002Fforums.fast.ai\u002Ft\u002Fmy-first-gold-medal\u002F54237)），有人获得了顶尖公司的录用通知（[详情](https:\u002F\u002Fforums.fast.ai\u002Ft\u002Fhow-has-your-journey-been-so-far-learners\u002F6480\u002F2)），还有人发表了研究论文（[详情](https:\u002F\u002Fui.adsabs.harvard.edu\u002Fabs\u002F2020EGUGA..2221465A\u002Fabstract)、[详情](http:\u002F\u002Fwww.ieomsociety.org\u002Fieom2020\u002Fpapers\u002F37.pdf)、[详情](https:\u002F\u002Fpubs.rsna.org\u002Fdoi\u002Fabs\u002F10.1148\u002Fryai.2019190113?journalCode=ai)）。例如，Isaac Dimitrovsky 曾告诉我们：“我玩机器学习已有几年，却始终不得要领……去年底参加了 fast.ai 第一部分课程后，我才真正开窍。”随后，他一举夺得享有盛誉的国际 RA2-DREAM 挑战赛冠军！他基于 fastai 库，开发了一种多阶段深度学习方法，用于评估类风湿性关节炎患者的放射学手足关节损伤程度。\n\n无论您是否具备技术或数学背景——当然，有也没关系！——我们都致力于让尽可能多的人轻松掌握深度学习。本课程唯一的先决条件是您会编程（一年左右的经验即可），最好使用 Python，并且至少修过高中数学课程。前三个章节特意以通俗易懂的方式编写，即使是高管、产品经理等非技术人员也能快速理解深度学习的核心要点；如果您属于这一群体，可以直接跳过这些章节中的代码部分。\n\n深度学习是一种利用多层神经网络来提取和转换数据的计算机技术，其应用场景涵盖从语音识别到动物图像分类等多个领域。许多人误以为要取得出色的深度学习成果，必须具备各种稀缺资源，但通过本课程的学习，您会发现这种观点并不正确。以下是一些“完全不需要”的所谓“必备条件”：\n\n| 误区（无需具备） | 真相 |\n|---|---|\n| 大量数学知识 | 高中数学水平已足够 |\n| 海量数据 | 我们曾见证仅用不到 50 个样本就创造纪录级效果 |\n| 昂贵的计算设备 | 即使是免费工具，也能完成最先进的深度学习任务 |\n\n深度学习兼具强大能力、高度灵活性和简洁性，因此我们认为它应当被广泛应用于各个学科领域，包括社会科学、自然科学、艺术、医学、金融、科学研究等等。以下是深度学习或深度学习主导的方法目前在全球范围内处于领先地位的数千项任务示例：\n\n- **自然语言处理 (NLP)**：问答系统、语音识别、文档摘要、文档分类、从文档中提取姓名、日期等信息、搜索提及特定概念的文章\n- **计算机视觉**：卫星与无人机影像解读（如灾害风险评估）、人脸识别、图像字幕生成、交通标志识别、自动驾驶车辆中的行人与车辆定位\n- **医学**：在 CT、MRI 和 X 光等影像中检测异常、病理切片特征计数、超声检查中测量指标、糖尿病视网膜病变诊断\n- **生物学**：蛋白质折叠、蛋白质分类、多项基因组学任务，例如肿瘤与正常组织测序、临床可干预基因突变分类、细胞分类、蛋白质间相互作用分析\n- **图像生成**：给黑白照片上色、提升图像分辨率、去除图像噪声、将照片转化为著名艺术家风格的艺术作品\n- **推荐系统**：网页搜索、商品推荐、首页内容布局\n- **游戏**：国际象棋、围棋、大多数 Atari 游戏以及许多即时战略游戏\n- **机器人技术**：处理难以定位（如透明、反光、缺乏纹理）或难以抓取的物体\n- **其他应用**：金融与物流预测、文本转语音等……\n\n## 我们是谁\n\n我们是 Sylvain Gugger 和 Jeremy Howard，您本次旅程的导师。我们是 fastai 的共同作者，而 fastai 正是您在本课程中将全程使用的软件。\n\nJeremy 从事机器学习的研究与教学已有约 30 年，早在 25 年前就开始使用神经网络。在此期间，他领导过多家以机器学习为核心的企业和项目，包括创立全球首家专注于深度学习与医疗的公司 Enlitic，并担任全球最大机器学习社区 Kaggle 的总裁兼首席科学家。此外，他还与 Rachel Thomas 博士共同创立了 fast.ai，也就是本课程的开发机构。\n\nSylvain 曾编写 10 册数学教材，覆盖法国高级数学课程的全部内容！如今，他任职于 Hugging Face，此前则在 fast.ai 担任研究员。\n\n我们非常重视教学。在本课程中，我们首先通过简单直观的工具，展示如何使用一个完整、可用且功能强大的先进深度学习模型来解决实际问题。随后，我们会逐步深入探讨这些工具的工作原理，以及构建这些工具的基础工具又是如何运作的，以此类推……我们始终采用实例教学，确保每个知识点都有清晰的情境和明确的目的，而不是一开始就进行抽象的符号运算。\n\n## 你将使用的软件\n\n在本课程中，你将会使用 [PyTorch](https:\u002F\u002Fpytorch.org\u002F) 和 [fastai](https:\u002F\u002Fdocs.fast.ai)。\n\n我们曾使用数十种不同的库和多种编程语言完成了数百个机器学习项目。在 fast.ai，我们已经用当今主流的深度学习和机器学习框架开设过多门课程。在决定未来课程、软件开发和研究都采用 PyTorch 之前，我们花了超过一千个小时对其进行测试。如今，PyTorch 已经成为全球发展最快的深度学习库，并且被用于顶级会议上的大多数研究论文。\n\nPyTorch 最适合作为底层基础库，提供更高层功能所需的基本操作。而 fastai 库则是最流行的在 PyTorch 基础之上添加这些高层功能的工具包。在本课程中，随着我们对深度学习基础的逐步深入，我们也会不断探索 fastai 的各个层次。本课程涵盖 fastai 库的 2.0 版本，这是一个从头开始重写的版本，提供了许多独特的功能。\n\n## 你将学到什么\n\n完成本课程后，你将掌握：\n\n- 如何训练能够达到当前最先进水平的模型，应用于以下领域：\n  - 计算机视觉，包括图像分类（例如按品种对宠物照片进行分类）以及图像定位与检测（例如找出图像中动物的位置）；\n  - 自然语言处理（NLP），包括文档分类（例如电影评论的情感分析）和语言建模；\n  - 表格数据（例如销售预测），涵盖类别型数据、连续型数据及混合型数据，还包括时间序列；\n  - 协同过滤（例如电影推荐）。\n- 如何将你的模型转化为 Web 应用并进行部署。\n- 深度学习模型的工作原理及其背后的机制，并学会如何利用这些知识来提升模型的准确性、速度和可靠性。\n- 实际应用中真正重要的最新深度学习技术。\n- 如何从零开始实现随机梯度下降算法及完整的训练循环。\n- 如何思考你的工作可能带来的伦理影响，以确保你的努力能够推动社会进步，同时避免技术被滥用造成伤害。\n\n以下是部分将要讲解的技术（如果你目前对这些术语还不太熟悉也不用担心——很快你就会全部掌握）：\n\n- 随机森林与梯度提升；\n- 线性变换与非线性激活函数；\n- 参数与激活值；\n- 随机初始化与迁移学习；\n- SGD、动量法、Adam 等优化器；\n- 卷积运算；\n- 批归一化；\n- Dropout 正则化；\n- 数据增强；\n- 权重衰减；\n- 图像分类与回归；\n- 实体嵌入与词嵌入；\n- 循环神经网络（RNN）；\n- 图像分割；\n- 以及更多内容。","# course20 (Practical Deep Learning for Coders) 快速上手指南\n\n本指南基于 fastai 和 PyTorch，旨在帮助开发者无需博士学位即可上手前沿的深度学习应用。\n\n## 环境准备\n\n本课程强烈建议**不要**在本地计算机配置环境（除非您精通 Linux 系统管理、GPU 驱动及 CUDA 配置），而是直接使用云端 Jupyter 服务器。这能确保所有课程代码（包括 Web GUI 功能）正常运行。\n\n### 推荐平台\n请从以下平台中选择一个启动环境：\n\n1.  **Gradient (首选)**\n    *   **特点**：完整的 Jupyter Notebook 体验，支持保存笔记和模型，课程所有功能（包括第 2 章的 Web 应用）均可用。\n    *   **注意**：免费服务器偶尔会过载导致无法连接。\n    *   **入口**：访问课程网站的 `\u002Fstart_gradient` 页面获取指引。\n\n2.  **Google Colab**\n    *   **特点**：谷歌提供的免费服务，普及率高。\n    *   **限制**：键盘快捷键与标准 Jupyter 不同；不支持创建 Web 应用 GUI（导致第 2 章部分内容无法运行）。\n    *   **入口**：访问课程网站的 `\u002Fstart_colab` 页面获取指引。\n\n3.  **Google Cloud (GCP)**\n    *   **特点**：提供完整的 Linux 服务器体验，极其可靠。\n    *   **注意**：高性能 GPU 费用较高，适合有进阶需求的用户。\n    *   **入口**：访问课程网站的 `\u002Fstart_gcp` 页面获取指引。\n\n### 前置知识\n*   **编程**：至少一年编程经验（推荐使用 Python）。\n*   **数学**：高中数学水平即可。\n\n## 安装步骤\n\n如果您选择使用推荐的云端平台（Gradient 或 Colab），**无需手动安装**任何依赖。这些环境已预装了课程所需的特定版本：\n\n*   **PyTorch**: 深度学习基础库。\n*   **fastai (v2)**: 基于 PyTorch 的高级封装库，本课程的核心工具。\n\n云端笔记本启动后，直接导入即可使用：\n\n```python\nfrom fastai.vision.all import *\nimport torch\n```\n\n> **注意**：若坚持在本地 Linux 环境部署，需自行处理 CUDA、cuDNN 及驱动兼容性，并参考 `docs.fast.ai` 安装最新版的 fastai v2 和 PyTorch，但这不在本快速指南推荐范围内。\n\n## 基本使用\n\n本课程的核心是通过交互式 Jupyter Notebook 学习。每个视频章节对应一个完整的 Notebook。\n\n### 1. 获取课程笔记\n课程所有章节均以交互式 Jupyter Notebook 形式提供。\n*   **在线阅读\u002F运行**：直接在上述推荐的云端平台中打开对应的课程笔记链接。\n*   **源码仓库**：所有笔记源码也可在 GitHub 上的 [fastbook](https:\u002F\u002Fgithub.com\u002Ffastai\u002Ffastbook) 仓库免费获取。\n\n### 2. 最简单的使用示例\n以下是一个使用 fastai 进行图像分类（识别宠物品种）的最小化代码示例，展示了如何加载数据、创建模型并进行训练：\n\n```python\nfrom fastai.vision.all import *\n\n# 1. 下载并解压示例数据集 (牛津宠物数据集)\npath = untar_data(URLs.PETS)\nfiles = get_image_files(path\u002F\"images\")\n\n# 2. 定义数据变换和数据加载器\n# 根据文件名中的数字判断类别 (例如: american_bulldog_1.jpg -> 类别 0)\ndef label_func(f): return f[0].isupper()\n\ndls = ImageDataLoaders.from_name_func(\n    path, files, valid_pct=0.2, seed=42,\n    label_func=label_func, item_tfms=Resize(224)\n)\n\n# 3. 创建并训练模型 (使用预训练的 ResNet34)\nlearn = vision_learner(dls, resnet34, metrics=error_rate)\nlearn.fine_tune(1)\n\n# 4. 进行预测\nimg = PILImage.create('test_image.jpg') # 替换为您的图片路径\npred_class, pred_idx, outputs = learn.predict(img)\nprint(f\"预测结果：{pred_class}, 置信度：{outputs[pred_idx].item():.4f}\")\n```\n\n### 3. 开始学习\n1.  进入课程网站，点击侧边栏的 **Lessons** 观看视频（支持中文字幕：点击 \"CC\" 并在设置中选择 \"简体中文\"）。\n2.  打开对应的 Jupyter Notebook，逐单元格运行代码并尝试修改参数进行实验。\n3.  遇到问题时，可访问 [fast.ai 论坛](https:\u002F\u002Fforums.fast.ai\u002Fc\u002Fpart1-v4\u002F46) 搜索解决方案或提问。","一位只有基础 Python 经验的电商数据分析师，试图为公司构建一个自动识别商品图片中缺陷的深度学习模型。\n\n### 没有 course20 时\n- 被复杂的数学公式和理论推导劝退，认为必须拥有博士学位才能入门深度学习，迟迟不敢动手写代码。\n- 在配置本地 GPU 环境、安装 CUDA 驱动和处理版本兼容性上耗费数周时间，导致项目尚未开始就已停滞。\n- 面对从零搭建神经网络架构的庞大工程量感到无从下手，难以将业务问题转化为具体的代码实现。\n- 缺乏系统的实践指导，遇到报错时只能在零散的技术博客中盲目搜索，效率极低且容易放弃。\n\n### 使用 course20 后\n- 遵循“先代码后理论”的理念，直接利用 fastai 库在几行代码内跑通第一个图像分类模型，迅速建立信心。\n- 直接使用课程推荐的 Gradient 或 Colab 云端 Jupyter 环境，无需任何本地配置即可立即开始训练高性能模型。\n- 通过书中配套的交互式 Notebook 逐步拆解任务，快速学会如何加载数据、微调预训练模型并部署应用。\n- 依托课程视频的可搜索字幕和活跃的官方论坛社区，能精准定位并解决如 Bing API 密钥获取等具体实操难题。\n\ncourse20 的核心价值在于它打破了学术高墙，让普通开发者无需深奥的数学背景也能高效落地真实的 AI 应用。","https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ffastai_course20_5ad69813.png","fastai","fast.ai","https:\u002F\u002Foss.gittoolsai.com\u002Favatars\u002Ffastai_50edd66d.png","",null,"fastdotai","https:\u002F\u002Ffast.ai","https:\u002F\u002Fgithub.com\u002Ffastai",[81,85,89,93,97,101,105],{"name":82,"color":83,"percentage":84},"Jupyter Notebook","#DA5B0B",83.8,{"name":86,"color":87,"percentage":88},"JavaScript","#f1e05a",10.8,{"name":90,"color":91,"percentage":92},"Python","#3572A5",3.3,{"name":94,"color":95,"percentage":96},"HTML","#e34c26",0.9,{"name":98,"color":99,"percentage":100},"CSS","#663399",0.7,{"name":102,"color":103,"percentage":104},"Makefile","#427819",0.3,{"name":106,"color":107,"percentage":108},"Shell","#89e051",0.1,840,303,"2026-04-11T20:47:53","Apache-2.0","Linux","强烈建议使用带 GPU 的云端服务器（如 Gradient, Colab, Google Cloud）。本地运行需具备 NVIDIA GPU、CUDA 驱动及相应的系统管理能力，具体型号和显存未说明。","未说明",{"notes":117,"python":118,"dependencies":119},"官方强烈不建议在个人电脑上运行本课程，除非用户非常精通 Linux 系统管理、GPU 驱动安装及 CUDA 配置。推荐直接使用 Gradient 或 Colab 等在线平台以避免环境配置问题。其中 Colab 存在部分功能限制（如不支持创建 Web GUI），Gradient 免费服务器可能过载。课程代码主要基于 Python，需具备至少一年编程经验。","未说明 (建议熟悉 Python)",[120,121,82],"fastai (v2)","PyTorch",[14],[124,125,126,127,128],"deep-learning","machine-learning","teaching","python","jupyter-notebook","2026-03-27T02:49:30.150509","2026-04-15T07:11:23.354196",[132,137,142,147,152,157,162],{"id":133,"question_zh":134,"answer_zh":135,"source_url":136},33924,"如何为 fast.ai 课程贡献代码或文档？第一步该做什么？","在开始之前，需要先安装 `nbdev` 包，因为贡献指南中提到的 `nbdev_install_git_hooks` 命令包含在该包中。安装后，在克隆的仓库目录下运行该命令以安装 git hooks，它会在每次提交和合并时自动清理 notebook 中多余的元数据并避免合并冲突。","https:\u002F\u002Fgithub.com\u002Ffastai\u002Fcourse20\u002Fissues\u002F71",{"id":138,"question_zh":139,"answer_zh":140,"source_url":141},33925,"在 Python 3 中运行 `search_images_ddg` 函数时报错 `'str' object has no attribute 'decode'` 怎么办？","这是因为在 Python 3 中字符串默认已经是解码状态，不需要再次调用 `.decode()`。解决方法是修改 `fastbook\u002F__init__.py` 文件，移除相关代码中的 `.decode()` 调用，因为底层的 `urlread` 函数已经处理了解码工作。","https:\u002F\u002Fgithub.com\u002Ffastai\u002Fcourse20\u002Fissues\u002F45",{"id":143,"question_zh":144,"answer_zh":145,"source_url":146},33926,"AWS SageMaker 部署模板启动失败或找不到 fastai kernel 如何解决？","这是由于 conda 环境更新依赖冲突导致的。解决方案是修改 CloudFormation 模板脚本：\n1. 在 OnCreate 脚本的 `echo \"Updating conda\"` 之后添加一行：`conda update --force-reinstall conda -y`。\n2. 移除 OnStart 脚本中更新 conda 的部分。\n3. 或者将 `pip install -r ...` 修改为 `pip install --ignore-installed -r ...` 以忽略已安装的包版本冲突。","https:\u002F\u002Fgithub.com\u002Ffastai\u002Fcourse20\u002Fissues\u002F60",{"id":148,"question_zh":149,"answer_zh":150,"source_url":151},33927,"Paperspace Gradient 的免费层级是否还能使用？文档似乎过时了。","Paperspace Gradient 仍然提供免费层级（Free Tier）。之前的文档可能未及时更新导致误解。用户可以参考 Paperspace 官方最新的入门教程获取设置指南，无需必须购买 Pro 套餐即可开始学习。","https:\u002F\u002Fgithub.com\u002Ffastai\u002Fcourse20\u002Fissues\u002F63",{"id":153,"question_zh":154,"answer_zh":155,"source_url":156},33928,"Google Colab 现在是否支持课程第 2 章中的交互式 widgets？","是的，Google Colab 现在已经支持 fast.ai 课程第 2 章中使用的交互式 widgets。之前存在的兼容性问题已解决，用户可以直接在 Colab 环境中正常运行包含这些组件的 Notebook。","https:\u002F\u002Fgithub.com\u002Ffastai\u002Fcourse20\u002Fissues\u002F37",{"id":158,"question_zh":159,"answer_zh":160,"source_url":161},33929,"为什么直接访问仓库中的 `start_colab` 或 `start_gradient` 文件会返回 404 错误？","因为这些文件是生成 course.fast.ai 官方网站文档的源文件，并不是设计用来直接在 GitHub 仓库页面查看或执行的。请访问正式的課程网站 (course.fast.ai) 查看渲染后的完整设置指南。","https:\u002F\u002Fgithub.com\u002Ffastai\u002Fcourse20\u002Fissues\u002F26",{"id":163,"question_zh":164,"answer_zh":165,"source_url":166},33930,"如何在贡献文档时避免内容被视为过度推广？","在提交关于第三方服务（如 iko.ai 等）的文档时，应专注于具体的操作指南（例如如何添加计算资源、如何在后台运行 Notebook 等），避免包含过多的营销性描述或功能罗列。确保内容对新学生实用且中立，维护者才会接受合并。","https:\u002F\u002Fgithub.com\u002Ffastai\u002Fcourse20\u002Fissues\u002F66",[]]