├── GoodArticle ├── 14 Design Patterns To Improve Your Convolutional Neural Networks.pdf ├── A Comprehensive Introduction to Different Types of Convolutions in Deep Learning.pdf ├── Bilateral Filtering: Theory and Applications.pdf ├── einsum满足你一切需要:深度学习中的爱因斯坦求和约定 - 知乎.pdf ├── 从SGD到NadaMax,十种优化算法原理及实现 - 知乎.pdf ├── 半小时学会 PyTorch Hook - 知乎.pdf ├── 移动平均:你知道的与你不知道的 - 知乎.pdf └── 迁移学习小册子-王晋东.pdf ├── README.md └── _config.yml /GoodArticle/14 Design Patterns To Improve Your Convolutional Neural Networks.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/lartpang/MyCollection/ec320b5c38ee8cd3db0f10d08f71d4e8afd5a80a/GoodArticle/14 Design Patterns To Improve Your Convolutional Neural Networks.pdf -------------------------------------------------------------------------------- /GoodArticle/A Comprehensive Introduction to Different Types of Convolutions in Deep Learning.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/lartpang/MyCollection/ec320b5c38ee8cd3db0f10d08f71d4e8afd5a80a/GoodArticle/A Comprehensive Introduction to Different Types of Convolutions in Deep Learning.pdf -------------------------------------------------------------------------------- /GoodArticle/Bilateral Filtering: Theory and Applications.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/lartpang/MyCollection/ec320b5c38ee8cd3db0f10d08f71d4e8afd5a80a/GoodArticle/Bilateral Filtering: Theory and Applications.pdf -------------------------------------------------------------------------------- /GoodArticle/einsum满足你一切需要:深度学习中的爱因斯坦求和约定 - 知乎.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/lartpang/MyCollection/ec320b5c38ee8cd3db0f10d08f71d4e8afd5a80a/GoodArticle/einsum满足你一切需要:深度学习中的爱因斯坦求和约定 - 知乎.pdf -------------------------------------------------------------------------------- /GoodArticle/从SGD到NadaMax,十种优化算法原理及实现 - 知乎.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/lartpang/MyCollection/ec320b5c38ee8cd3db0f10d08f71d4e8afd5a80a/GoodArticle/从SGD到NadaMax,十种优化算法原理及实现 - 知乎.pdf -------------------------------------------------------------------------------- /GoodArticle/半小时学会 PyTorch Hook - 知乎.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/lartpang/MyCollection/ec320b5c38ee8cd3db0f10d08f71d4e8afd5a80a/GoodArticle/半小时学会 PyTorch Hook - 知乎.pdf -------------------------------------------------------------------------------- /GoodArticle/移动平均:你知道的与你不知道的 - 知乎.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/lartpang/MyCollection/ec320b5c38ee8cd3db0f10d08f71d4e8afd5a80a/GoodArticle/移动平均:你知道的与你不知道的 - 知乎.pdf -------------------------------------------------------------------------------- /GoodArticle/迁移学习小册子-王晋东.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/lartpang/MyCollection/ec320b5c38ee8cd3db0f10d08f71d4e8afd5a80a/GoodArticle/迁移学习小册子-王晋东.pdf -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- 1 | # My Collection of Repos 2 | 3 | **NOTE**: 后续更新可见,在语雀上写文档还是容易些。 4 | 5 | ![image](https://user-images.githubusercontent.com/26847524/68121367-8ffa9800-ff42-11e9-80cd-38aeb860bb98.png) 6 | 7 | ## 声明 8 | 9 | 由于几篇文章写的内容对于本人而言非常有用,但是为了防止原作“消失”的情况,这里将其进行了打印,保存成了pdf,存放在这里的`GoodArticle`文件夹中,如果原作者觉得不妥,请联系我,我会删除。保护版权,人人有责! 10 | 11 | * 迁移学习小册子-王晋东: https://zhuanlan.zhihu.com/p/35352154 12 | * Bilateral Filtering: Theory and Applications: https://www.researchgate.net/publication/220427978_Bilateral_Filtering_Theory_and_Applications 13 | * 半小时学会 PyTorch Hook - 尹相楠的文章 - 知乎:https://zhuanlan.zhihu.com/p/75054200 14 | * 14 DESIGN PATTERNS TO IMPROVE YOUR CONVOLUTIONAL NEURAL NETWORKS:https://www.topbots.com/14-design-patterns-improve-convolutional-neural-network-cnn-architecture/ 15 | * 移动平均:你知道的与你不知道的 - 石川的文章 - 知乎:https://zhuanlan.zhihu.com/p/38276041 16 | * A Comprehensive Introduction to Different Types of Convolutions in Deep Learning: https://towardsdatascience.com/a-comprehensive-introduction-to-different-types-of-convolutions-in-deep-learning-669281e58215 17 | * einsum满足你一切需要:深度学习中的爱因斯坦求和约定 - 知乎: https://zhuanlan.zhihu.com/p/44954540 18 | * 从SGD到NadaMax,十种优化算法原理及实现 - 知乎: https://zhuanlan.zhihu.com/p/81020717 19 | 20 | ## 基础 21 | 22 | + CS229:https://github.com/afshinea/stanford-cs-229-machine-learning 23 | + transferlearning:https://github.com/jindongwang/transferlearning 24 | + 数据挖掘:https://github.com/lyltj2010/DataMining 25 | + 机器学习:https://github.com/allmachinelearning/MachineLearning 26 | + 论文翻译: https://github.com/SnailTyan/deep-learning-papers-translation 27 | + 综述论文: https://github.com/mlreview/machine-learning-surveys 28 | + 深度学习500问:https://github.com/scutan90/DeepLearning-500-questions 29 | + 深度学习面试宝典(含数学、机器学习、深度学习、计算机视觉、自然语言处理和SLAM等方向):https://github.com/amusi/Deep-Learning-Interview-Book 30 | + zouxy09博客原创性博文导航:https://blog.csdn.net/zouxy09/article/details/14222605 31 | + DeepLearning ON GPU:https://deeplearningongpu.readthedocs.io/en/latest/ 32 | + ML Records in 1110 Lab of BUPT:https://github.com/wmn7/ML_Practice 33 | + 各种计算机领域资料的合集:https://github.com/ty4z2008/Qix 34 | * 如何理解深度学习分布式训练中的large batch size与learning rate的关系? - 谭旭的回答 - 知乎:https://www.zhihu.com/question/64134994/answer/216895968 35 | + 数据增强 36 | * This is a list of awesome methods about data augmentation.: https://github.com/CrazyVertigo/awesome-data-augmentation 37 | * A collection of awesome things about mixed sample data augmentation: https://github.com/JasonZhang156/awesome-mixed-sample-data-augmentation 38 | 39 | ### 数学 40 | 41 | + L1和L2范数:http://www.chioka.in/differences-between-l1-and-l2-as-loss-function-and-regularization/ 42 | + 想要算一算Wasserstein距离?这里有一份PyTorch实战:https://www.jiqizhixin.com/articles/19031102 43 | + Gumbel-Softmax Trick(将不可数的argmax采样)和Gumbel分布:https://www.cnblogs.com/initial-h/p/9468974.html 44 | > Gumbel Trick 是一种从离散分布取样的方法,它的形式可以允许我们定义一种可微分的,离散分布的近似取样,这种取样方式不像「干脆以各类概率值的概率向量替代取样」这么粗糙,也不像直接取样一样不可导(因此没办法应对可能的 bp )。如何理解Gumbel-Max trick? - 曹恭泽的回答 - 知乎 45 | 46 | ### 弱监督、半监督以及蒸馏等 47 | 48 | + 让机器“一叶知秋”:弱监督视觉语义分割:https://zhuanlan.zhihu.com/p/37488849 49 | + 介绍了CVPR19的一篇使用蒸馏处理语义分割的论文:https://segmentfault.com/a/1190000018493751 50 | + 半监督深度学习小节:https://zhuanlan.zhihu.com/p/33196506 51 | + https://github.com/iBelieveCJM/pseudo_label-pytorch 52 | + 半监督深度学习又小结之Consistency Regularization:https://zhuanlan.zhihu.com/p/46893709 53 | + Attention算法调研(弱监督语义分割):https://zhuanlan.zhihu.com/p/53218967 54 | 55 | ### 多任务学习 56 | 57 | + 多任务学习Multi-task Learning(MTL)概述:https://blog.csdn.net/zaf0516/article/details/90380732 58 | + 一片综述:https://arxiv.org/pdf/1706.05098.pdf 59 | + 多任务学习(Multi-Task Learning, MTL):https://blog.csdn.net/laolu1573/article/details/78205180# 60 | + 多任务深度学习的三个经验教训:https://www.leiphone.com/news/201902/SlFB1OlWd6U8ubJj.html 61 | + Multi-Task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semantics:https://arxiv.org/pdf/1705.07115.pdf 62 | 63 | ## CV 64 | 65 | * Awesome-Deblurring: https://github.com/subeeshvasu/Awesome-Deblurring 66 | * awesome-gan-for-medical-imaging: https://github.com/xinario/awesome-gan-for-medical-imaging 67 | * How to Train a GAN? Tips and tricks to make GANs work: https://github.com/soumith/ganhacks 68 | * Awesome Database Learning: https://github.com/pingcap/awesome-database-learning 69 | * ppwwyyxx/GroupNorm-reproduce: https://github.com/ppwwyyxx/GroupNorm-reproduce 70 | * awesome-gan-for-medical-imaging: https://github.com/xinario/awesome-gan-for-medical-imaging 71 | * 自己的学习历程,重点包括各种好玩的图像处理算法、运动捕捉、机器学习: https://github.com/1165048017/BlogLearning 72 | 73 | ### 行人重识别(ReID) 74 | 75 | + 较为简单直观的一份行人重识别的代码**教程**: https://github.com/layumi/Person_reID_baseline_pytorch 76 | + 比较全面完整的一份代码:https://github.com/L1aoXingyu/reid_baseline 77 | + 行人重识别(ReID)**概述**,将该任务的关键内容介绍了一下:https://blog.csdn.net/weixin_41427758/article/details/81188164 78 | + 行人重识别重要数据集`DukeMTMC-reID`的一个简单介绍,了解一个任务的数据处理,看数据集的结构是比较直接的一个方法:https://blog.csdn.net/ctwy291314/article/details/83544142 79 | 80 | ## 书籍 81 | 82 | ### Coding 83 | 84 | * 综合 85 | * 设计数据密集型应用 - 中文翻译: https://github.com/Vonng/ddia 86 | * Computer Science from the Bottom Up: https://www.bottomupcs.com/index.xhtml 87 | * Mining Social Media: http://socialdata.site/ 88 | * 重新审视《GOTO 语句被认为有害》: https://www.emon100.me/goto-translation/ 89 | * GitHub中文排行榜,帮助你发现高分优秀中文项目、更高效地吸收国人的优秀经验成果;榜单每周更新一次,敬请关注!Resources: 90 | * Editor 91 | * 笨方法学Vimscript: https://www.kancloud.cn/kancloud/learn-vimscript-the-hard-way 92 | * VSCode: https://geek-docs.com/vscode 93 | * Lisp 94 | * build-your-own-lisp: https://ksco.gitbooks.io/build-your-own-lisp/content/ 95 | * Rust 96 | * Rust 简明教程: https://geektutu.com/post/quick-rust.html 97 | * Python 98 | * Python 最佳实践指南 2018: https://learnku.com/docs/python-guide/2018 99 | * Python最佳实践指南!: https://pythonguidecn.readthedocs.io/zh/latest/index.html 100 | * What the f*ck Python! 🐍(中文翻译): https://github.com/leisurelicht/wtfpython-cn 101 | * Python - 100天从新手到大师:https://github.com/jackfrued/Python-100-Days 102 | * :stars: 来自一位 Pythonista 的编程经验分享,内容涵盖编码技巧、最佳实践与思维模式等方面:https://github.com/piglei/one-python-craftsman 103 | * :stars: Comprehensive Python Cheatsheet: 104 | * Python深入06 Python的内存管理:https://www.cnblogs.com/vamei/p/3232088.html 105 | * 10种检测Python程序运行时间、CPU和内存占用的方法:https://www.jb51.net/article/63244.htm 106 | * Golang: 107 | * Building a BitTorrent client from the ground up in Go: https://blog.jse.li/posts/torrent/ 108 | * Go 101: https://go101.org/article/101.html 109 | * Go 高级教程: https://github.com/chai2010/advanced-go-programming-book 110 | * 《Go语法树入门——开启自制编程语言和编译器之旅》(开源免费图书/Go语言进阶/掌握抽象语法树/Go语言AST/LLVM/LLIR/凹语言):https://github.com/chai2010/go-ast-book 111 | * 7days-golang: https://github.com/geektutu/7days-golang 112 | * Go 语言简明教程: https://geektutu.com/post/quick-golang.html 113 | * Bash 114 | * dylanaraps/pure-bash-bible:https://github.com/dylanaraps/pure-bash-bible 115 | * Algorithm 116 | * 手把手撕LeetCode题目,扒各种算法套路的裤子:https://github.com/labuladong/fucking-algorithm 117 | * 小浩算法是我在疫情期间完成的一部图解算法题典! 目前共完成 140+ 道高频面试算法题目,总计 30w 字,全部采用漫画图解的方式,简单易懂,适合初中级读者:https://github.com/geekxh/hello-algorithm 118 | 119 | ### 线性代数 120 | 121 | + Matrix Analysis and Applied Linear Algebra:http://www.cse.zju.edu.cn/eclass/attachments/2015-10/01-1446085870-145420.pdf 122 | + https://book.douban.com/subject/1479316/ 123 | + Introduction to Linear Algebra(4th):http://www.math.hcmus.edu.vn/~bxthang/Linear%20algebra%20and%20its%20applications.pdf 124 | + https://book.douban.com/subject/3582335/ 125 | + https://github.com/golang101/golang101 126 | 127 | ### ML&DL 128 | 129 | + 吴恩达新书: http://nooverfit.com/wp/wp-content/uploads/2018/10/Ng-MLY01-12.pdf 130 | 131 | ## 工具 132 | 133 | ### 文档 134 | 135 | * TOML中文文档:https://github.com/LongTengDao/TOML/ 136 | 137 | ### 多媒体 138 | 139 | * 【FFmepg】学习音视频知识,整理资料,编写技术手册:https://github.com/feixiao/ffmpeg 140 | 141 | ### Python 142 | 143 | + 将程序转化为命令行接口:https://github.com/google/python-fire 144 | 145 | ### PyTorch 146 | 147 | + 统计模型运算量和参数量的pytorch-OpCounter:https://github.com/Lyken17/pytorch-OpCounter 148 | + 一些PyTorch的技巧,但是似乎有些旧了:https://github.com/kevinzakka/pytorch-goodies 149 | + PyTorch Autograd:Understanding the heart of PyTorch’s magic:https://towardsdatascience.com/pytorch-autograd-understanding-the-heart-of-pytorchs-magic-2686cd94ec95 150 | + CNN可视化: 151 | - https://github.com/MisaOgura/flashtorch 152 | - https://github.com/utkuozbulak/pytorch-cnn-visualizations 153 | - Ways for Visualizing Convolutional Networks:https://buptldy.github.io/2016/09/25/2016-09-25-cnn_vis/ 154 | - CNN-Visualization:https://github.com/scutan90/CNN-Visualization 155 | - 谷歌的新CNN特征可视化方法,构造出一个华丽繁复的新世界:https://www.leiphone.com/news/201711/aNw8ZjqMuqvygzlz.html 156 | - How to visualize convolutional features in 40 lines of code:https://towardsdatascience.com/how-to-visualize-convolutional-features-in-40-lines-of-code-70b7d87b0030 157 | + 中文翻译:https://www.jishuwen.com/d/2Lik 158 | - 【深度学习系列】CNN模型的可视化:https://www.cnblogs.com/charlotte77/p/8343700.html 159 | + 高级封装: 160 | - https://github.com/blue-season/pywarm 161 | + 训练提速: 162 | - 简单两步加速PyTorch里的Dataloader - 巽二的文章 - 知乎:https://zhuanlan.zhihu.com/p/68191407 163 | - 给训练踩踩油门——Pytorch加速数据读取 - MrTian的文章 - 知乎:https://zhuanlan.zhihu.com/p/80695364 164 | - https://www.cnblogs.com/king-lps/p/10936374.html 165 | + Image Test Time Augmentation with PyTorch!: https://github.com/qubvel/ttach 166 | 167 | ### TF 168 | 169 | + tensorflow教程:https://github.com/open-source-for-science/TensorFlow-Course#why-use-tensorflow 170 | 171 | ### Editor 172 | 173 | + SublimeText:本身没啥问题,而且随着版本的更新,原本在linux上的各种问题基本上也被解决,但是有个关键的问题,在使用GPU跑代码的时候,ST会特别卡顿,只好舍弃,转投VSCode了 174 | + VSCode:微软的产品,还不错,性能问题逐渐得到优化,是肉眼可见的提升与改进,颜值可以很高 175 | + Vim:终端修改文件必备,偶尔看看文件,或者远程通过ssh使用Vim写写代码,但是用的不是很痛快,终究还是被PyCharm这样的IDE惯坏了 176 | + TeXmacs:一款对于数学编辑极其方便的工具,从FTP下载顺道可以下载对应的guile(a Scheme implementation):ftp://ftp.texmacs.org/TeXmacs/tmftp/Linux/ 177 | 178 | ### IDE 179 | 180 | + PyCharm:写Python必备,前提是电脑配置不要太拖后腿,时常卡顿,但是功能强大又贴心,实在无法替代 181 | + wingide:流畅是流畅,但是一来要激活,二来在我的Ubuntu上无法输入中文,而且创建项目的诡异流程让我怀疑这玩意是怎么想的:(,弃之 182 | 183 | ### Linux 184 | 185 | + ubuntu16.04安装NIVIDIA显卡驱动,cuda8.0,cuDNN6.0以及基于Anaconda安装Tensorflow-GPU:https://blog.csdn.net/pursuit_zhangyu/article/details/79362128 186 | + 【解决】Ubuntu安装NVIDIA驱动(咨询NVIDIA工程师的解决方案):https://blog.csdn.net/u012759136/article/details/53355781#commentBox 187 | + [专业亲测]Ubuntu16.04安装Nvidia显卡驱动(cuda)--解决你的所有困惑:https://blog.csdn.net/ghw15221836342/article/details/79571559 188 | + 最全面解析 Ubuntu 16.04 安装nvidia驱动 以及各种错误:https://blog.csdn.net/u014561933/article/details/79958017 189 | + ubuntu安装nvidia显卡驱动:https://blog.csdn.net/acelove40/article/details/69257574 190 | + Ubuntu 15.10 使用 Xorg.conf 修改分辨率:https://segmentfault.com/a/1190000004510095 191 | 192 | ### Windows 193 | 194 | * windows的下载镜像链接藏得好深,速度可真慢,根本不给你下载的机会啊,隔壁`I Tell You`都是用`ed2k`链接,好用的工具也只有迅雷,但迅雷(~~浏览器?!!~~)又不好好做下载,不让人喜欢,现在对于一般的下载感觉也限速了的样子,Windows重装的梦想看来难以实现了:https://www.microsoft.com/zh-cn/software-download/windows10ISO 195 | * 不要手动下载镜像,可以使用前面页面中提供的安装介质自动下载镜像,速度又快~:https://zhuanlan.zhihu.com/p/78326370 196 | 197 | ## 文字编辑 198 | 199 | * 中英混排中的标点符号问题:https://thetype.com/2018/02/14211/ 200 | 201 | ### 字体 202 | 203 | * 中英文兼顾的编程字体:https://github.com/be5invis/Sarasa-Gothic 204 | * 非常棒的英文编程字体(要不是因为不支持中文,我就用这个了,这个字体偏宽一些):https://github.com/tonsky/FiraCode 205 | 206 | ### Web 207 | 208 | * Placeholder.com – The Free Image Placeholder Service Favoured By Designers:https://placeholder.com/ 209 | -------------------------------------------------------------------------------- /_config.yml: -------------------------------------------------------------------------------- 1 | theme: jekyll-theme-midnight --------------------------------------------------------------------------------