收集一些文章的

归一化和标准化，权重初始化，batchnormal

Batch Normalization详解 - shine-lee - 博客园 (cnblogs.com)https://www.cnblogs.com/shine-lee/p/11989612.html

Python 语言参考手册 — Python 3.9.7 文档https://docs.python.org/zh-cn/3/reference/index.html

Python3 教程 | 菜鸟教程 (runoob.com)https://www.runoob.com/python3/python3-tutorial.html

神经网络理解http://playground.lieweiai.cn/#activation=tanh&batchSize=10&dataset=circle®Dataset=reg-plane&learningRate=0.03®ularizationRate=0&noise=0&networkShape=4,2&seed=0.68148&showTestData=false&discretize=false&percTrainData=50&x=true&y=true&xTimesY=false&xSquared=false&ySquared=false&cosX=false&sinX=false&cosY=false&sinY=false&collectStats=false&problem=classification&initZero=false&hideText=false

NumPy 中文https://www.numpy.org.cn/

(8条消息) Numpy知识点大总结_蹦跶跶-CSDN博客https://blog.csdn.net/weixin_44515978/article/details/107889577

The latest in Machine Learning | Papers With Codehttps://paperswithcode.com/

深入理解图像的卷积 - 知乎 (zhihu.com)https://zhuanlan.zhihu.com/p/128107523 An Intuitive Explanation of Convolutional Neural Networks – the data science blog (ujjwalkarn.me)https://ujjwalkarn.me/2016/08/11/intuitive-explanation-convnets/

如何理解卷积神经网络（CNN）中的卷积和池化？ - 知乎 (zhihu.com)https://www.zhihu.com/question/49376084 为什么选交叉熵作为分类问题的损失函数 - 知乎 (zhihu.com)https://zhuanlan.zhihu.com/p/81956135

【教程】学会Git玩转Github【全】_哔哩哔哩_bilibilihttps://www.bilibili.com/video/BV1Xx411m7kn?p=3&spm_id_from=pageDriver

一个小时学会Git - 张果 - 博客园 (cnblogs.com)https://www.cnblogs.com/best/p/7474442.html

机器学习中正则化项L1和L2的直观理解_小平子的专栏-CSDN博客_l1和l2正则化https://blog.csdn.net/jinping_shi/article/details/52433975

[编程-Pytorch-数据加载提速]-2 - 知乎 (zhihu.com)https://zhuanlan.zhihu.com/p/261222303

Dataset Zoo · deepinsight/insightface Wiki · GitHub（一些人脸数据集）https://github.com/deepinsight/insightface/wiki/Dataset-Zoo

 MSCOCO api详解 —— Keypoints - 知乎 (zhihu.com)https://zhuanlan.zhihu.com/p/135355550

(一) COCO Python API - 使用篇_Right here, Code is Magic-CSDN博客_cocohttps://blog.csdn.net/gzj2013/article/details/82385164

coco/pycocoDemo.ipynb（coco使用例子）https://github.com/dengdan/coco/blob/master/PythonAPI/pycocoDemo.ipynb

【论文解读】Yolo三部曲解读——Yolov3 - 知乎 (zhihu.com)https://zhuanlan.zhihu.com/p/76802514

label smoothing(标签平滑)学习笔记 - 知乎 (zhihu.com)https://zhuanlan.zhihu.com/p/116466239GitHub - wudashuo/yolov5: YOLOv5 汉化版，保持官方同步更新https://github.com/wudashuo/yolov5 显著性检测数据集（一些分割数据集）https://blog.csdn.net/studyeboy/article/details/102383922

 模型剪裁教程 · Pytorch 中文文档 (apachecn.org)https://pytorch.apachecn.org/docs/1.7/54.html

 pytorch模型剪枝学习笔记 - 尚码园 (shangmayuan.com)https://www.shangmayuan.com/a/2d5129a988154737a3611b46.html

 知识蒸馏是什么？一份入门随笔 - 知乎 (zhihu.com)https://zhuanlan.zhihu.com/p/90049906

State Farm Distracted Driver Detection | Kagglehttps://www.kaggle.com/c/state-farm-distracted-driver-detection/data 【深度学习笔记】关键点检测——标签（Ground Truth）构建_李响-CSDN博客_深度学习关键点https://blog.csdn.net/weixin_43838785/article/details/103585656

NTU-RGBD骨架数据分析_bairw_Bella的博客-CSDN博客https://blog.csdn.net/bairw_Bella/article/details/109028137

（原）人体姿态识别alphapose - darkknightzh - 博客园 (cnblogs.com)https://www.cnblogs.com/darkknightzh/p/12150171.html  人体姿态估计Alphapose配置安装教程（GPU,超详细，亲测有效！）_fengfeng，Z的博客-CSDN博客_alphapose安装https://blog.csdn.net/qq_38522972/article/details/82993216

人体姿态识别研究综述（详细归纳！）_fengfeng，Z的博客-CSDN博客_姿态识别https://blog.csdn.net/qq_38522972/article/details/82953477 人体姿态识别研究综述（详细归纳！）_fengfeng，Z的博客-CSDN博客_姿态识别一，人体姿态识别数据集1，2D数据集：LSP地址：http://sam.johnson.io/research/lsp.html样本数：2K关节点个数：14全身，单人FLIC地址：https://bensapp.github.io/flic-dataset.html样本数：2W关节点个数：9全身，单人MPII地址：http://human-pos...https://blog.csdn.net/qq_38522972/article/details/82953477

Deep Learning Monitor （深度学习最新论文跟踪）https://deeplearn.org/

目标检测 YOLOv5 转ncnn移动端部署_flyfish-CSDN博客https://flyfish.blog.csdn.net/article/details/116604907

GitHub - nihui/ncnn-android-yolov5: The YOLOv5 object detection android examplehttps://github.com/nihui/ncnn-android-yolov5

GitHub - bubbliiiing/yolov4-tiny-pytorch: 这是一个YoloV4-tiny-pytorch的源码，可以用于训练自己的模型。https://github.com/bubbliiiing/yolov4-tiny-pytorch#%E8%AE%AD%E7%BB%83%E6%AD%A5%E9%AA%A4 Pytorch 搭建自己的YoloV4目标检测平台（Bubbliiiing 深度学习 教程）_哔哩哔哩_bilibilihttps://www.bilibili.com/video/BV1Q54y1D7vj?p=16

 CVPR 2020：华为GhostNet，超越谷歌MobileNet，已开源 - 知乎 (zhihu.com)https://zhuanlan.zhihu.com/p/109325275

 GitHub - deepcam-cn/yolov5-face: YOLO5Face: Why Reinventing a Face Detector (https://arxiv.org/abs/2105.12931)https://github.com/deepcam-cn/yolov5-face

 Pytorch转NCNN的流程记录 - 知乎 (zhihu.com)https://zhuanlan.zhihu.com/p/124294444

一键转换 Caffe, ONNX, TensorFlow 到 NCNN, MNN, Tengine (convertmodel.com)https://convertmodel.com/#outputFormat=tnn GitHub - CrazyVertigo/awesome-data-augmentation: This is a list of awesome methods about data augmentation.https://github.com/CrazyVertigo/awesome-data-augmentation

OpenCV—python 颜色空间（RGB，HSV，Lab）与 颜色直方图_wsp_1138886114的博客-CSDN博客_opencv 颜色https://blog.csdn.net/wsp_1138886114/article/details/80660014  Pytorch 中的数据增强方式最全解释_Leong Blog-CSDN博客https://blog.csdn.net/LSJ944830401/article/details/102917336?ops_request_misc=%257B%2522request%255Fid%2522%253A%2522162797385616780357289069%2522%252C%2522scm%2522%253A%252220140713.130102334..%2522%257D&request_id=162797385616780357289069&biz_id=0&utm_medium=distribute.pc_search_result.none-task-blog-2~all~sobaiduend~default-2-102917336.first_rank_v2_pc_rank_v29&utm_term=RandomAffine&spm=1018.2226.3001.4187

大白话 Generalized Focal Loss - 知乎 (zhihu.com)https://zhuanlan.zhihu.com/p/147691786  大白话 Generalized Focal Loss - 知乎 (zhihu.com)https://zhuanlan.zhihu.com/p/147691786

albumentations 数据增强工具的使用 - 知乎 (zhihu.com)https://zhuanlan.zhihu.com/p/107399127/

图像增强库Albumentations使用总结_AI浩-CSDN博客https://blog.csdn.net/hhhhhhhhhhwwwwwwwwww/article/details/116460996

无监督难分样本挖掘改进目标检测 - 知乎 (zhihu.com)https://zhuanlan.zhihu.com/p/174057800