{"id":19023978,"url":"https://github.com/harleyszhang/dl_note","last_synced_at":"2025-05-15T23:08:25.571Z","repository":{"id":65222452,"uuid":"572592804","full_name":"harleyszhang/dl_note","owner":"harleyszhang","description":"深度学习系统笔记，包含深度学习数学基础知识、神经网络基础部件详解、深度学习炼丹策略、模型压缩算法详解。","archived":false,"fork":false,"pushed_at":"2025-04-27T03:47:05.000Z","size":197720,"stargazers_count":467,"open_issues_count":0,"forks_count":63,"subscribers_count":4,"default_branch":"main","last_synced_at":"2025-04-27T04:28:39.047Z","etag":null,"topics":["activation-functions","cnn","convolutional-neural-networks","cpp","deep-learning","inference-framework","loss-functions","machine-learning","python","pytorch"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/harleyszhang.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2022-11-30T15:53:14.000Z","updated_at":"2025-04-27T03:47:10.000Z","dependencies_parsed_at":"2023-01-15T16:15:28.598Z","dependency_job_id":"b51d668a-2f16-45b5-92e3-964a4550c20f","html_url":"https://github.com/harleyszhang/dl_note","commit_stats":{"total_commits":386,"total_committers":3,"mean_commits":"128.66666666666666","dds":"0.031088082901554404","last_synced_commit":"c6b8a4ea1c61c12dcaa41049581e14a0bda912b3"},"previous_names":["harleyszhang/dl_note"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/harleyszhang%2Fdl_note","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/harleyszhang%2Fdl_note/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/harleyszhang%2Fdl_note/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/harleyszhang%2Fdl_note/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/harleyszhang","download_url":"https://codeload.github.com/harleyszhang/dl_note/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":254436949,"owners_count":22070947,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["activation-functions","cnn","convolutional-neural-networks","cpp","deep-learning","inference-framework","loss-functions","machine-learning","python","pytorch"],"created_at":"2024-11-08T20:34:15.599Z","updated_at":"2025-05-15T23:08:19.191Z","avatar_url":"https://github.com/harleyszhang.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"- [项目概述](#项目概述)\n- [一，数学和编程基础专栏](#一数学和编程基础专栏)\n- [二，神经网络基础部件](#二神经网络基础部件)\n- [三，经典卷积神经网络模型](#三经典卷积神经网络模型)\n- [四，深度学习炼丹](#四深度学习炼丹)\n- [五，深度学习模型压缩](#五深度学习模型压缩)\n- [六，模型推理部署](#六模型推理部署)\n- [七，进阶课程](#七进阶课程)\n- [参考资料](#参考资料)\n\n## 项目概述\n\n本仓库项目是个人总结的计算机视觉和大语言模型学习笔记，包含深度学习基础知识、神经网络基础部件详解、深度学习炼丹策略、深度学习模型压缩算法、深度学习推理框架代码解析及动手实战。\n\n`LLM` 基础及推理优化的专栏笔记请参考[llm_note](https://github.com/HarleysZhang/llm_note)仓库。\n\n## 一，数学和编程基础专栏\n\n- [深度学习数学基础-概率与信息论](./1-math_ml_basic/深度学习数学基础-概率与信息论.md)\n- [深度学习基础-机器学习基本原理](./1-math_ml_basic/深度学习基础-机器学习基本原理.md)\n- [随机梯度下降法的数学基础](./1-math_ml_basic/随机梯度下降法的数学基础.md)\n- [Python 编程思维导航](./1-math_ml_basic/python_learn_xmind)\n\n## 二，神经网络基础部件\n\n1，**神经网络基础部件**：\n\n1. [神经网络基础部件-卷积层详解](./2-deep_learning_basic/神经网络基础部件-卷积层详解.md)\n2. [神经网络基础部件-BN 层详解](./2-deep_learning_basic/神经网络基础部件-BN层详解.md)\n3. [神经网络基础部件-激活函数详解](./2-deep_learning_basic/神经网络基础部件-激活函数详解.md)\n\n2，**深度学习基础**：\n\n- [反向传播与梯度下降详解](2-deep_learning_basic/反向传播与梯度下降详解.md)\n- [深度学习基础-参数初始化详解](./2-deep_learning_basic/深度学习基础-参数初始化详解.md)\n- [深度学习基础-损失函数详解](./2-deep_learning_basic/深度学习基础-损失函数详解.md)\n- [深度学习基础-优化算法详解](./2-deep_learning_basic/深度学习基础-优化算法详解.md)\n\n## 三，经典卷积神经网络模型\n\n**1，卷积神经网络的经典 backbone**：\n\n- [ResNet网络详解](3-classic_backbone/ResNet网络详解.md)\n- [DenseNet 网络详解](3-classic_backbone/DenseNet论文解读.md)\n- [ResNetv2 网络详解](3-classic_backbone/ResNetv2论文解读.md)\n- [经典 backbone 网络总结](3-classic_backbone/经典backbone总结.md)\n\n**2，轻量级网络详解**：\n\n- [MobileNetv1论文详解](3-classic_backbone/efficient_cnn/MobileNetv1论文详解.md)\n- [ShuffleNetv2论文详解](3-classic_backbone/efficient_cnn/ShuffleNetv2论文详解.md)\n- [RepVGG论文详解](3-classic_backbone/efficient_cnn/RepVGG论文详解.md)\n- [CSPNet论文详解](3-classic_backbone/efficient_cnn/CSPNet论文详解.md)\n- [VoVNet论文解读](3-classic_backbone/efficient_cnn/VoVNet论文解读.md)\n- [轻量级模型设计总结](5-model_compression/模型压缩-轻量化网络总结.md)\n\n## 四，深度学习炼丹\n\n1. [深度学习炼丹-数据标准化](./4-deep_learning_alchemy/深度学习炼丹-数据标准化.md)\n2. [深度学习炼丹-数据增强](./4-deep_learning_alchemy/深度学习炼丹-数据增强.md)\n3. [深度学习炼丹-不平衡样本的处理](./4-deep_learning_alchemy/深度学习炼丹-不平衡样本的处理.md)\n4. [深度学习炼丹-超参数设定](./4-deep_learning_alchemy/深度学习炼丹-超参数调整.md)\n5. [深度学习炼丹-正则化策略](./4-deep_learning_alchemy/深度学习炼丹-正则化策略.md)\n\n## 五，深度学习模型压缩\n\n1. [深度学习模型压缩算法综述](./5-model_compression/深度学习模型压缩方法概述.md)\n2. [模型压缩-轻量化网络设计与部署总结](./5-model_compression/模型压缩-轻量化网络详解.md)\n3. [模型压缩-剪枝算法详解](./5-model_compression/模型压缩-剪枝算法详解.md)\n4. [模型压缩-知识蒸馏详解](./5-model_compression/模型压缩-知识蒸馏详解.md)\n5. [模型压缩-量化算法详解](./5-model_compression/模型压缩-量化算法概述.md)\n\n## 六，模型推理部署\n\n1，模型推理部署：\n\n- [卷积神经网络复杂度分析](./6-model_deploy/卷积神经网络复杂度分析.md)\n- [模型压缩部署概述](./6-model_deploy/模型压缩部署概述.md)\n- [矩阵乘法详解](./6-model_deploy/卷积算法优化.md)\n- [模型推理加速技巧-融合卷积和BN层](./6-model_deploy/模型推理加速技巧-融合卷积和BN层.md)\n\n2，`ncnn` 框架源码解析：\n\n- [ncnn 源码解析-sample 运行](5-model_deploy/ncnn源码解析-sample运行.md)\n- [ncnn 源码解析-Net 类](5-model_deploy/ncnn源码解析-Net类.md)\n\n3，异构计算\n\n1. 移动端异构计算：`neon` 编程\n2. GPU 端异构计算：`cuda` 编程，比如 `gemm` 算法解析与优化\n\n## 七，进阶课程\n\n1，推荐几个比较好的深度学习模型压缩与加速的仓库和课程资料：\n\n1. [神经网络基本原理教程](https://github.com/microsoft/ai-edu/blob/master/%E5%9F%BA%E7%A1%80%E6%95%99%E7%A8%8B/A2-%E7%A5%9E%E7%BB%8F%E7%BD%91%E7%BB%9C%E5%9F%BA%E6%9C%AC%E5%8E%9F%E7%90%86/%E7%AC%AC8%E6%AD%A5%20-%20%E5%8D%B7%E7%A7%AF%E7%A5%9E%E7%BB%8F%E7%BD%91%E7%BB%9C/17.1-%E5%8D%B7%E7%A7%AF%E7%9A%84%E5%89%8D%E5%90%91%E8%AE%A1%E7%AE%97%E5%8E%9F%E7%90%86.md)\n2. [AI-System](https://microsoft.github.io/AI-System/): 深度学习系统，主要从底层方向讲解深度学习系统等原理、加速方法、矩阵成乘加计算等。\n3. [pytorch-deep-learning](https://github.com/mrdbourke/pytorch-deep-learning)：很好的 pytorch 深度学习教程。\n\n2，一些笔记好的博客链接：\n\n- [The Illustrated Transformer](http://jalammar.github.io/illustrated-transformer/): 国内比较好的博客大都参考这篇文章。\n- [C++ 并发编程（从C++11到C++17）](https://paul.pub/cpp-concurrency/): 不错的 C++ 并发编程教程。\n- [What are Diffusion Models?](https://lilianweng.github.io/posts/2021-07-11-diffusion-models/)\n- [annotated_deep_learning_paper_implementations](https://github.com/labmlai/annotated_deep_learning_paper_implementations)\n\n3，最后，持续高质量创作不易，有 `5` 秒空闲时间的，**可以扫码关注我的公众号-嵌入式视觉**，记录 CV 算法工程师成长之路，分享技术总结、读书笔记和个人感悟。\n\u003e 公众号不会写标题党文章，也不输出给大家带来的焦虑的内容！\n\n![qcode](images/others/qcode.png)\n\n4，Star History Chart：\n\n[![Star History Chart](https://api.star-history.com/svg?repos=HarleysZhang/deep_learning_system\u0026type=Date)](https://star-history.com/#HarleysZhang/deep_learning_system\u0026Date)\n\n## 参考资料\n\n- 《深度学习》\n- 《机器学习》\n- 《动手学深度学习》\n- [《机器学习系统：设计和实现》](https://openmlsys.github.io/index.html)\n- [《AI-EDU》](https://ai-edu.openai.wiki/%E5%9F%BA%E7%A1%80%E6%95%99%E7%A8%8B/index.html)\n- [《AI-System》](https://github.com/microsoft/AI-System/tree/main/Textbook)\n- [《PyTorch_tutorial_0.0.5_余霆嵩》](https://github.com/TingsongYu/PyTorch_Tutorial)\n- [《动手编写深度学习推理框架 Planer》](https://github.com/Image-Py/planer)\n- [distill：知识精要和在线可视化](https://distill.pub/)\n- [LLVM IR入门指南](https://github.com/Evian-Zhang/llvm-ir-tutorial)\n- [nanoPyC](https://github.com/vesuppi/nanoPyC/tree/master)\n- [ClassifyTemplate](https://github.com/Yale1417/ClassifyTemplate)\n- [pytorch-classification](https://github.com/bearpaw/pytorch-classification)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fharleyszhang%2Fdl_note","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fharleyszhang%2Fdl_note","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fharleyszhang%2Fdl_note/lists"}