# 深度学习编译 **Repository Path**: liushuo_ustc/DeepLearningCompiling ## Basic Information - **Project Name**: 深度学习编译 - **Description**: 深度学习编译资料共享 - **Primary Language**: Unknown - **License**: LGPL-2.1 - **Default Branch**: master - **Homepage**: None - **GVP Project**: No ## Statistics - **Stars**: 0 - **Forks**: 4 - **Created**: 2023-07-30 - **Last Updated**: 2023-07-30 ## Categories & Tags **Categories**: Uncategorized **Tags**: None ## README ### 文档参考资料列表 :exclamation: :exclamation: :exclamation: 说明：文档参考资料列表整理自根目录下的各个章节文件夹，由于各章节参考资料存在交叉以及便于阅读整理，建议及时上传资料至相应文件夹并按 **如下格式更新此列表中的链接** 以及添加资料简要的介绍等内容 #### 01深度学习简介 - 1、[循环神经网络研究综述_杨丽](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/01%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%AE%80%E4%BB%8B/%E8%83%8C%E6%99%AF_%E8%83%8C%E6%99%AF%E5%BE%AA%E7%8E%AF%E7%A5%9E%E7%BB%8F%E7%BD%91%E7%BB%9C%E7%A0%94%E7%A9%B6%E7%BB%BC%E8%BF%B0_%E6%9D%A8%E4%B8%BD.pdf) - 2、[深度学习相关研究综述_张军阳](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/01%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%AE%80%E4%BB%8B/%E8%83%8C%E6%99%AF_%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%9B%B8%E5%85%B3%E7%A0%94%E7%A9%B6%E7%BB%BC%E8%BF%B0_%E5%BC%A0%E5%86%9B%E9%98%B3.pdf) - 3、[深度学习技术和平台发展综述_于佃海](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/01%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%AE%80%E4%BB%8B/%E8%83%8C%E6%99%AF_%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E6%8A%80%E6%9C%AF%E5%92%8C%E5%B9%B3%E5%8F%B0%E5%8F%91%E5%B1%95%E7%BB%BC%E8%BF%B0_%E4%BA%8E%E4%BD%83%E6%B5%B7.pdf) - 4、[图神经网络技术研究综述_李甜甜](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/01%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%AE%80%E4%BB%8B/%E8%83%8C%E6%99%AF_%E5%9B%BE%E7%A5%9E%E7%BB%8F%E7%BD%91%E7%BB%9C%E6%8A%80%E6%9C%AF%E7%A0%94%E7%A9%B6%E7%BB%BC%E8%BF%B0_%E6%9D%8E%E7%94%9C%E7%94%9C.pdf) - 5、[卷积神经网络研究综述_周飞燕](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/01%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%AE%80%E4%BB%8B/%E8%83%8C%E6%99%AF_%E5%8D%B7%E7%A7%AF%E7%A5%9E%E7%BB%8F%E7%BD%91%E7%BB%9C%E7%A0%94%E7%A9%B6%E7%BB%BC%E8%BF%B0_%E5%91%A8%E9%A3%9E%E7%87%95.pdf) - 6、[神经网络软硬件协同加速关键技术_王佩琪](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/01%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%AE%80%E4%BB%8B/%E7%A1%AC%E4%BB%B6_%E7%A5%9E%E7%BB%8F%E7%BD%91%E7%BB%9C%E8%BD%AF%E7%A1%AC%E4%BB%B6%E5%8D%8F%E5%90%8C%E5%8A%A0%E9%80%9F%E5%85%B3%E9%94%AE%E6%8A%80%E6%9C%AF_%E7%8E%8B%E4%BD%A9%E7%90%AA.pdf) - 7、[计算机视觉代码示例](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/01%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%AE%80%E4%BB%8B/%E8%AE%A1%E7%AE%97%E6%9C%BA%E8%A7%86%E8%A7%89%E7%A4%BA%E4%BE%8B%E4%BB%A3%E7%A0%81.docx) 地址：https://www.jianshu.com/p/a46191daacfc - 8、[目标检测代码示例](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/01%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%AE%80%E4%BB%8B/%E7%9B%AE%E6%A0%87%E6%A3%80%E6%B5%8B%E7%A4%BA%E4%BE%8B%E4%BB%A3%E7%A0%81.zip) 地址：https://github.com/qfs1980398040/pytorch-Single-target-detection-Minions - 9、[自然语言处理代码示例](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/01%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%AE%80%E4%BB%8B/%E8%87%AA%E7%84%B6%E8%AF%AD%E8%A8%80%E5%A4%84%E7%90%86%E7%A4%BA%E4%BE%8B%E4%BB%A3%E7%A0%81.zip) 地址：https://github.com/graykode/nlp-tutorial - 10、[语音识别代码示例](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/01%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%AE%80%E4%BB%8B/%E8%AF%AD%E9%9F%B3%E8%AF%86%E5%88%AB%E7%A4%BA%E4%BE%8B%E4%BB%A3%E7%A0%81.zip) 地址：https://github.com/xxbb1234021/speech_recognition - 11、[王秉睿. 神经网络专用编程语言[D].中国科学技术大学,2019](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/01%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%AE%80%E4%BB%8B/Domain%20Specif...ural%20Networks_%E7%8E%8B%E7%A7%89%E7%9D%BF.pdf) - 12、[图神经网络前沿进展与应用_吴博](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/01%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%AE%80%E4%BB%8B/%E5%9B%BE%E7%A5%9E%E7%BB%8F%E7%BD%91%E7%BB%9C%E5%89%8D%E6%B2%BF%E8%BF%9B%E5%B1%95%E4%B8%8E%E5%BA%94%E7%94%A8_%E5%90%B4%E5%8D%9A.pdf) - 13、[深度学习在计算机视觉领域的应用进展_曾子力](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/01%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%AE%80%E4%BB%8B/%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E5%9C%A8%E8%AE%A1%E7%AE%97%E6%9C%BA%E8%A7%86%E8%A7%89%E9%A2%86%E5%9F%9F%E7%9A%84%E5%BA%94%E7%94%A8%E8%BF%9B%E5%B1%95_%E6%9B%BE%E5%AD%90%E5%8A%9B.pdf) - 14、[面向深度学习硬件加速器的网络编译工具设计_严天炜](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/01%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%AE%80%E4%BB%8B/%E9%9D%A2%E5%90%91%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%A1%AC%E4%BB%B6%E5%8A%A0%E9%80%9F%E5%99%A8%E7%9A%84%E7%BD%91%E7%BB%9C%E7%BC%96%E8%AF%91%E5%B7%A5%E5%85%B7%E8%AE%BE%E8%AE%A1_%E4%B8%A5%E5%A4%A9%E7%82%9C.pdf) - 15、[TensorFlow内核剖析](https://github.com/horance-liu/tensorflow-internals/blob/master/tensorflow-internals.pdf) #### 02-03深度学习编译系统设计&深度学习硬件平台 - 1、[背景_面向实时应用的深度学习研究综述_张政馗](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/01%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%AE%80%E4%BB%8B/%E8%83%8C%E6%99%AF_%E9%9D%A2%E5%90%91%E5%AE%9E%E6%97%B6%E5%BA%94%E7%94%A8%E7%9A%84%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%A0%94%E7%A9%B6%E7%BB%BC%E8%BF%B0_%E5%BC%A0%E6%94%BF%E9%A6%97.pdf) - 2、[神经网络软硬件协同加速关键技术_王佩琪](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/01%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%AE%80%E4%BB%8B/%E7%A1%AC%E4%BB%B6_%E7%A5%9E%E7%BB%8F%E7%BD%91%E7%BB%9C%E8%BD%AF%E7%A1%AC%E4%BB%B6%E5%8D%8F%E5%90%8C%E5%8A%A0%E9%80%9F%E5%85%B3%E9%94%AE%E6%8A%80%E6%9C%AF_%E7%8E%8B%E4%BD%A9%E7%90%AA.pdf) - 3、[深度神经网络加速器体系结构概述_陈怡然](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/01%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%AE%80%E4%BB%8B/%E7%A1%AC%E4%BB%B6_%E6%B7%B1%E5%BA%A6%E7%A5%9E%E7%BB%8F%E7%BD%91%E7%BB%9C%E5%8A%A0%E9%80%9F%E5%99%A8%E4%BD%93%E7%B3%BB%E7%BB%93%E6%9E%84%E6%A6%82%E8%BF%B0_%E9%99%88%E6%80%A1%E7%84%B6.pdf) - 4、[深度学习计算平台发展综述_郭乔进](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/01%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%AE%80%E4%BB%8B/%E7%A1%AC%E4%BB%B6_%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E8%AE%A1%E7%AE%97%E5%B9%B3%E5%8F%B0%E5%8F%91%E5%B1%95%E7%BB%BC%E8%BF%B0_%E9%83%AD%E4%B9%94%E8%BF%9B.pdf) - 5、[基于NVDLA的深度学习推断芯片研究_周高峰](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/01%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%AE%80%E4%BB%8B/%E7%A1%AC%E4%BB%B6_%E5%9F%BA%E4%BA%8ENVDLA%E7%9A%84%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E6%8E%A8%E6%96%AD%E8%8A%AF%E7%89%87%E7%A0%94%E7%A9%B6_%E5%91%A8%E9%AB%98%E5%B3%B0.pdf) - 6、[基于FPGA的深度可分离卷积神经网络加速器设计研究_詹宏毅](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/01%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%AE%80%E4%BB%8B/%E7%A1%AC%E4%BB%B6_%E5%9F%BA%E4%BA%8EFPGA%E7%9A%84%E6%B7%B1%E5%BA%A6%E5%8F%AF%E5%88%86%E7%A6%BB%E5%8D%B7%E7%A7%AF%E7%A5%9E%E7%BB%8F%E7%BD%91%E7%BB%9C%E5%8A%A0%E9%80%9F%E5%99%A8%E8%AE%BE%E8%AE%A1%E7%A0%94%E7%A9%B6_%E8%A9%B9%E5%AE%8F%E6%AF%85.pdf) - 7、[人工智能加速体系结构综述_陈正博](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/01%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%AE%80%E4%BB%8B/%E7%A1%AC%E4%BB%B6_%E4%BA%BA%E5%B7%A5%E6%99%BA%E8%83%BD%E5%8A%A0%E9%80%9F%E4%BD%93%E7%B3%BB%E7%BB%93%E6%9E%84%E7%BB%BC%E8%BF%B0_%E9%99%88%E6%AD%A3%E5%8D%9A.pdf) - 8、[基于FPGA的通用深度卷积神经网络加速器设计_管兆康](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/01%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%AE%80%E4%BB%8B/%E7%A1%AC%E4%BB%B6_%E5%9F%BA%E4%BA%8EFPGA%E7%9A%84%E9%80%9A%E7%94%A8%E6%B7%B1%E5%BA%A6%E5%8D%B7%E7%A7%AF%E7%A5%9E%E7%BB%8F%E7%BD%91%E7%BB%9C%E5%8A%A0%E9%80%9F%E5%99%A8%E8%AE%BE%E8%AE%A1_%E7%AE%A1%E5%85%86%E5%BA%B7.pdf) - 9、[一种混合架构的高可靠性深度神经网络加速器_王乾龙](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/01%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%AE%80%E4%BB%8B/%E7%A1%AC%E4%BB%B6_%E4%B8%80%E7%A7%8D%E6%B7%B7%E5%90%88%E6%9E%B6%E6%9E%84%E7%9A%84%E9%AB%98%E5%8F%AF%E9%9D%A0%E6%80%A7%E6%B7%B1%E5%BA%A6%E7%A5%9E%E7%BB%8F%E7%BD%91%E7%BB%9C%E5%8A%A0%E9%80%9F%E5%99%A8_%E7%8E%8B%E4%B9%BE%E9%BE%99.pdf) - 10、[FPGA加速器深度卷积神经网络优化计算方法_梁修壮](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/01%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%AE%80%E4%BB%8B/%E7%A1%AC%E4%BB%B6_FPGA%E5%8A%A0%E9%80%9F%E5%99%A8%E6%B7%B1%E5%BA%A6%E5%8D%B7%E7%A7%AF%E7%A5%9E%E7%BB%8F%E7%BD%91%E7%BB%9C%E4%BC%98%E5%8C%96%E8%AE%A1%E7%AE%97%E6%96%B9%E6%B3%95_%E6%A2%81%E4%BF%AE%E5%A3%AE.pdf) - 11、[DNN加速器技术发展及航空计算系统应用展望_赵一煊](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/01%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%AE%80%E4%BB%8B/%E7%A1%AC%E4%BB%B6_DNN%E5%8A%A0%E9%80%9F%E5%99%A8%E6%8A%80%E6%9C%AF%E5%8F%91%E5%B1%95%E5%8F%8A%E8%88%AA%E7%A9%BA%E8%AE%A1%E7%AE%97%E7%B3%BB%E7%BB%9F%E5%BA%94%E7%94%A8%E5%B1%95%E6%9C%9B_%E8%B5%B5%E4%B8%80%E7%85%8A.pdf) - 12、[基于龙芯平台的深度学习算子库的构建与优化_尹宁](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/01%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%AE%80%E4%BB%8B/%E7%A1%AC%E4%BB%B6+%E6%A1%86%E6%9E%B6_%E5%9F%BA%E4%BA%8E%E9%BE%99%E8%8A%AF%E5%B9%B3%E5%8F%B0%E7%9A%84%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%AE%97%E5%AD%90%E5%BA%93%E7%9A%84%E6%9E%84%E5%BB%BA%E4%B8%8E%E4%BC%98%E5%8C%96_%E5%B0%B9%E5%AE%81.pdf) - 13、[Designing Deep Learning Hardware Accelerator and efficiency evaluation](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/02%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%A1%AC%E4%BB%B6%E5%B9%B3%E5%8F%B0/Designing%20Deep%20Learning%20Hardware%20Accelerator%20and%20efficiency%20evaluation.pdf) - 14、[2022中国人工智能芯片行业研究报告](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/02%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%A1%AC%E4%BB%B6%E5%B9%B3%E5%8F%B0/2022%E4%B8%AD%E5%9B%BD%E4%BA%BA%E5%B7%A5%E6%99%BA%E8%83%BD%E8%8A%AF%E7%89%87%E8%A1%8C%E4%B8%9A%E7%A0%94%E7%A9%B6%E6%8A%A5%E5%91%8A.pdf) - 15、智能计算系统 [视频公开课](https://forum.cambricon.com/index.php?m=content&c=index&a=show&catid=154&id=232) [slide](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/02%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%A1%AC%E4%BB%B6%E5%B9%B3%E5%8F%B0/%E7%AC%AC%E5%85%AD%E7%AB%A0%20%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E5%A4%84%E7%90%86%E5%99%A8%E5%8E%9F%E7%90%86.pdf) - 16、[深度神经网络 FPGA 设计进展、实现与展望_焦李成](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/02%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%A1%AC%E4%BB%B6%E5%B9%B3%E5%8F%B0/%E6%B7%B1%E5%BA%A6%E7%A5%9E%E7%BB%8F%E7%BD%91%E7%BB%9C%20FPGA%20%E8%AE%BE%E8%AE%A1%E8%BF%9B%E5%B1%95%E3%80%81%E5%AE%9E%E7%8E%B0%E4%B8%8E%E5%B1%95%E6%9C%9B_%E7%84%A6%E7%A4%BC%E6%88%90.pdf) - 17、[人工智能芯片技术白皮书（2018）](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/02%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%A1%AC%E4%BB%B6%E5%B9%B3%E5%8F%B0/%E4%BA%BA%E5%B7%A5%E6%99%BA%E8%83%BD%E8%8A%AF%E7%89%87%E6%8A%80%E6%9C%AF%E7%99%BD%E7%9A%AE%E4%B9%A6%EF%BC%882018%EF%BC%89.pdf) - 18、[人工智能芯片研究报告（2018）](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/02%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%A1%AC%E4%BB%B6%E5%B9%B3%E5%8F%B0/aichip%E4%BA%BA%E5%B7%A5%E6%99%BA%E8%83%BD%E8%8A%AF%E7%89%87%E7%A0%94%E7%A9%B6%E6%8A%A5%E5%91%8A2018.pdf) - 19、[深度学习相关研究综述_张军阳：深度学习相关加速技术（2016）](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/01%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%AE%80%E4%BB%8B/%E8%83%8C%E6%99%AF_%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%9B%B8%E5%85%B3%E7%A0%94%E7%A9%B6%E7%BB%BC%E8%BF%B0_%E5%BC%A0%E5%86%9B%E9%98%B3.pdf) - 20、[System_for_AI Microsoft_01-02课程介绍&人工智能系统概述](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/System_for_AI%20Microsoft/01-02%E8%AF%BE%E7%A8%8B%E4%BB%8B%E7%BB%8D&%E4%BA%BA%E5%B7%A5%E6%99%BA%E8%83%BD%E7%B3%BB%E7%BB%9F%E6%A6%82%E8%BF%B0.pdf) - 21、[System_for_AI Microsoft_03深度神经网络计算框架基础](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/System_for_AI%20Microsoft/03%E6%B7%B1%E5%BA%A6%E7%A5%9E%E7%BB%8F%E7%BD%91%E7%BB%9C%E8%AE%A1%E7%AE%97%E6%A1%86%E6%9E%B6%E5%9F%BA%E7%A1%80.pdf) - 22、深入理解TensorFlow架构设计与实现原理_彭靖田[PDF01](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/1_P_%E6%B7%B1%E5%85%A5%E7%90%86%E8%A7%A3TensorFlow%E6%9E%B6%E6%9E%84%E8%AE%BE%E8%AE%A1%E4%B8%8E%E5%AE%9E%E7%8E%B0%E5%8E%9F%E7%90%86_%E5%BD%AD%E9%9D%96%E7%94%B0.pdf) [PDF02](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/198_P_%E6%B7%B1%E5%85%A5%E7%90%86%E8%A7%A3TensorFlow%E6%9E%B6%E6%9E%84%E8%AE%BE%E8%AE%A1%E4%B8%8E%E5%AE%9E%E7%8E%B0%E5%8E%9F%E7%90%86_%E5%BD%AD%E9%9D%96%E7%94%B0.pdf) -------------------------------------------深度学习编译系统设计---------------------------------------------------- - 1、[Deep Learning Systems: Algorithms, Compilers, and Processors for Large-Scale Production](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/Deep%20Learning%20Systems%20Algorithms%20Compilers%20and%20Processors%20for%20Large-Scale%20Production.pdf) - 2、[An_In-depth_Comparison_of_Compilers_for_Deep_Neural_Networks_on_Hardware](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/An_In-depth_Comparison_of_Compilers_for_Deep_Neural_Networks_on_Hardware.pdf) - 3、[The Deep Learning Compiler: A Comprehensive Survey by Mingzhen Li et al., TPDS 2020](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/07%E4%B8%AD%E9%97%B4%E8%A1%A8%E7%A4%BA/%E5%8F%82%E8%80%83%E8%B5%84%E6%96%99/The%20Deep%20Learning%20Compiler%20A%20Comprehensive%20Survey.pdf) - 4、[深度学习框架研究及初步实现_孙振](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/01%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%AE%80%E4%BB%8B/%E6%A1%86%E6%9E%B6_%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E6%A1%86%E6%9E%B6%E7%A0%94%E7%A9%B6%E5%8F%8A%E5%88%9D%E6%AD%A5%E5%AE%9E%E7%8E%B0_%E5%AD%99%E6%8C%AF.pdf) - 5、[智能遥感深度学习框架与模型设计_龚健雅](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/01%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%AE%80%E4%BB%8B/%E6%A1%86%E6%9E%B6_%E6%99%BA%E8%83%BD%E9%81%A5%E6%84%9F%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E6%A1%86%E6%9E%B6%E4%B8%8E%E6%A8%A1%E5%9E%8B%E8%AE%BE%E8%AE%A1_%E9%BE%9A%E5%81%A5%E9%9B%85.pdf) - 6、[新一代深度学习框架研究_于璠](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/01%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%AE%80%E4%BB%8B/%E6%96%B0%E4%B8%80%E4%BB%A3%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E6%A1%86%E6%9E%B6%E7%A0%94%E7%A9%B6_%E4%BA%8E%E7%92%A0.pdf) - 7、[基于TVM的移动端垃圾分类辅助识别系统_华林泉](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/01%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%AE%80%E4%BB%8B/%E5%9F%BA%E4%BA%8ETVM%E7%9A%84%E7%A7%BB%E5%8A%A8%E7%AB%AF%E5%9E%83%E5%9C%BE%E5%88%86%E7%B1%BB%E8%BE%85%E5%8A%A9%E8%AF%86%E5%88%AB%E7%B3%BB%E7%BB%9F_%E5%8D%8E%E6%9E%97%E6%B3%89.pdf) - 8、[基于CNN加速器的深度学习编译器设计与实现_张芳芳](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/01%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%AE%80%E4%BB%8B/%E5%9F%BA%E4%BA%8ECNN%E5%8A%A0%E9%80%9F%E5%99%A8%E7%9A%84%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E5%99%A8%E8%AE%BE%E8%AE%A1%E4%B8%8E%E5%AE%9E%E7%8E%B0_%E5%BC%A0%E8%8A%B3%E8%8A%B3.pdf) - 9、[AI 框架发展白皮书2022年](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/AI%20%E6%A1%86%E6%9E%B6%E5%8F%91%E5%B1%95%E7%99%BD%E7%9A%AE%E4%B9%A62022%E5%B9%B4.pdf) - 10、[王秉睿,兰慧盈,陈云霁.深度学习编程框架[J].大数据,2018,4(04):56-63](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/01%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%AE%80%E4%BB%8B/Programming%20f...ng%20algorithms_%E7%8E%8B%E7%A7%89%E7%9D%BF.pdf) - 11、[MLSys: The New Frontier of Machine Learning Systems](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/MLSys%20The%20New%20Frontier%20of%20Machine%20Learning%20Systems.pdf) - 12、[Jittor: a novel deep learning framework with meta-operators and unified graph execution](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/Jittor_%20a%20nov...aph%20execution_Shi-Min%20HU.pdf) - 13、[智能计算系统_陈云霁](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/%E6%99%BA%E8%83%BD%E8%AE%A1%E7%AE%97%E7%B3%BB%E7%BB%9F.pdf) - 14、[AI框架基础ZOMI_01AI框架基础概述](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/AI%E6%A1%86%E6%9E%B6%E5%9F%BA%E7%A1%80ZOMI/01AI%E6%A1%86%E6%9E%B6%E5%9F%BA%E7%A1%80%E6%A6%82%E8%BF%B0.pdf) - 15、[AI框架基础ZOMI_02AI框架作用](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/AI%E6%A1%86%E6%9E%B6%E5%9F%BA%E7%A1%80ZOMI/02AI%E6%A1%86%E6%9E%B6%E4%BD%9C%E7%94%A8.pdf) - 16、[AI框架基础ZOMI_03AI框架发展](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/AI%E6%A1%86%E6%9E%B6%E5%9F%BA%E7%A1%80ZOMI/03AI%E6%A1%86%E6%9E%B6%E5%8F%91%E5%B1%95.pdf) - 17、[编译器ZOMI_01课程概述](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/%E7%BC%96%E8%AF%91%E5%99%A8ZOMI/01%E8%AF%BE%E7%A8%8B%E6%A6%82%E8%BF%B0.pdf) - 18、[编译器ZOMI_02开源编译器的发展](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/%E7%BC%96%E8%AF%91%E5%99%A8ZOMI/02%E5%BC%80%E6%BA%90%E7%BC%96%E8%AF%91%E5%99%A8%E7%9A%84%E5%8F%91%E5%B1%95.pdf) - 19、[编译器ZOMI_03GCC编译过程和原理](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/%E7%BC%96%E8%AF%91%E5%99%A8ZOMI/03GCC%E7%BC%96%E8%AF%91%E8%BF%87%E7%A8%8B%E5%92%8C%E5%8E%9F%E7%90%86.pdf) - 20、[编译器ZOMI_04LLVM设计架构](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/%E7%BC%96%E8%AF%91%E5%99%A8ZOMI/04LLVM%E8%AE%BE%E8%AE%A1%E6%9E%B6%E6%9E%84.pdf) - 21、[编译器ZOMI_05LLVM IR详解](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/%E7%BC%96%E8%AF%91%E5%99%A8ZOMI/05LLVM%20IR%E8%AF%A6%E8%A7%A3.pdf) - 22、[编译器ZOMI_06LLVM前端和优化层](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/%E7%BC%96%E8%AF%91%E5%99%A8ZOMI/06LLVM%E5%89%8D%E7%AB%AF%E5%92%8C%E4%BC%98%E5%8C%96%E5%B1%82.pdf) - 23、[编译器ZOMI_07LLVM后端代码生成](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/%E7%BC%96%E8%AF%91%E5%99%A8ZOMI/07LLVM%E5%90%8E%E7%AB%AF%E4%BB%A3%E7%A0%81%E7%94%9F%E6%88%90.pdf) - 24、[编译器ZOMI_08为什么需要AI编译器](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/%E7%BC%96%E8%AF%91%E5%99%A8ZOMI/08%E4%B8%BA%E4%BB%80%E4%B9%88%E9%9C%80%E8%A6%81AI%E7%BC%96%E8%AF%91%E5%99%A8.pdf) - 25、[编译器ZOMI_09AI编译器的发展阶段](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/%E7%BC%96%E8%AF%91%E5%99%A8ZOMI/09AI%E7%BC%96%E8%AF%91%E5%99%A8%E7%9A%84%E5%8F%91%E5%B1%95%E9%98%B6%E6%AE%B5.pdf) - 26、[编译器ZOMI_10AI编译器的通用架构](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/%E7%BC%96%E8%AF%91%E5%99%A8ZOMI/10AI%E7%BC%96%E8%AF%91%E5%99%A8%E7%9A%84%E9%80%9A%E7%94%A8%E6%9E%B6%E6%9E%84.pdf) - 27、[编译器ZOMI_11AI编译器的挑战与思考](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/%E7%BC%96%E8%AF%91%E5%99%A8ZOMI/11AI%E7%BC%96%E8%AF%91%E5%99%A8%E7%9A%84%E6%8C%91%E6%88%98%E4%B8%8E%E6%80%9D%E8%80%83.pdf) - 28、[System_for_AI Microsoft_01-02课程介绍&人工智能系统概述](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/System_for_AI%20Microsoft/01-02%E8%AF%BE%E7%A8%8B%E4%BB%8B%E7%BB%8D&%E4%BA%BA%E5%B7%A5%E6%99%BA%E8%83%BD%E7%B3%BB%E7%BB%9F%E6%A6%82%E8%BF%B0.pdf) - 29、[System_for_AI Microsoft_04矩阵运算与计算机框架基础](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/System_for_AI%20Microsoft/04%E7%9F%A9%E9%98%B5%E8%BF%90%E7%AE%97%E4%B8%8E%E8%AE%A1%E7%AE%97%E6%9C%BA%E4%BD%93%E7%B3%BB%E7%BB%93%E6%9E%84.pdf) - 30、[System_for_AI Microsoft_11自动机器学习系统](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/System_for_AI%20Microsoft/11%E8%87%AA%E5%8A%A8%E6%9C%BA%E5%99%A8%E5%AD%A6%E4%B9%A0%E7%B3%BB%E7%BB%9F.pdf) - 31、[System_for_AI Microsoft_12强化学习系统](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/System_for_AI%20Microsoft/12%E5%BC%BA%E5%8C%96%E5%AD%A6%E4%B9%A0%E7%B3%BB%E7%BB%9F.pdf) - 32、[System_for_AI Microsoft_13安全与隐私](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/System_for_AI%20Microsoft/13%E5%AE%89%E5%85%A8%E4%B8%8E%E9%9A%90%E7%A7%81.pdf) - 33、[System_for_AI Microsoft_14利用人工智能来优化计算机系统](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/System_for_AI%20Microsoft/14%E5%88%A9%E7%94%A8%E4%BA%BA%E5%B7%A5%E6%99%BA%E8%83%BD%E6%9D%A5%E4%BC%98%E5%8C%96%E8%AE%A1%E7%AE%97%E6%9C%BA%E7%B3%BB%E7%BB%9F.pdf) #### 04编程接口 - 1、[FreeTensor: A Free-Form DSL with Holistic Optimizations for Irregular Tensor Programs by Shizhi Tang et al., PLDI 2022](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/04%E7%BC%96%E7%A8%8B%E6%8E%A5%E5%8F%A3/freetensor.pdf) - 2、[AI框架基础ZOMI_04AI框架编程范式](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/AI%E6%A1%86%E6%9E%B6%E5%9F%BA%E7%A1%80ZOMI/04%E7%BC%96%E7%A8%8B%E8%8C%83%E5%BC%8F.pdf) #### 05前端（类型系统与静态分析、计算图生成） - 1、[面向类型推导的Python类型标注分析_马洪跃](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/05%E7%B1%BB%E5%9E%8B%E7%B3%BB%E7%BB%9F%E4%B8%8E%E9%9D%99%E6%80%81%E5%88%86%E6%9E%90/Python%20Type%20A...pe%20Inferrence_%E9%A9%AC%E6%B4%AA%E8%B7%83.pdf) - 2、[Python静态类型分析及其应用_董天聪](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/05%E7%B1%BB%E5%9E%8B%E7%B3%BB%E7%BB%9F%E4%B8%8E%E9%9D%99%E6%80%81%E5%88%86%E6%9E%90/Static%20Type%20A...ng%20for%20Python_%E8%91%A3%E5%A4%A9%E8%81%AA.pdf) - 3、[浅谈Python中的可变与不可变数据类型_陈玲](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/05%E7%B1%BB%E5%9E%8B%E7%B3%BB%E7%BB%9F%E4%B8%8E%E9%9D%99%E6%80%81%E5%88%86%E6%9E%90/Variable%20and%20...pes%20in%20Python_%E9%99%88%E7%8E%B2.pdf) - 4、[一种Python外部函数的静态类型推断方法_张昱](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/05%E7%B1%BB%E5%9E%8B%E7%B3%BB%E7%BB%9F%E4%B8%8E%E9%9D%99%E6%80%81%E5%88%86%E6%9E%90/%E4%B8%80%E7%A7%8DPython%E5%A4%96%E9%83%A8%E5%87%BD%E6%95%B0%E7%9A%84%E9%9D%99%E6%80%81%E7%B1%BB%E5%9E%8B%E6%8E%A8%E6%96%AD%E6%96%B9%E6%B3%95%E5%8F%8A%E7%B3%BB%E7%BB%9F_%E5%BC%A0%E6%98%B1.pdf) - 5、[一种基于类型标注的Python程序类型推导方法_陈林](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/05%E7%B1%BB%E5%9E%8B%E7%B3%BB%E7%BB%9F%E4%B8%8E%E9%9D%99%E6%80%81%E5%88%86%E6%9E%90/%E4%B8%80%E7%A7%8D%E5%9F%BA%E4%BA%8E%E7%B1%BB%E5%9E%8B%E6%A0%87%E6%B3%A8%E7%9A%84Python%E7%A8%8B%E5%BA%8F%E7%B1%BB%E5%9E%8B%E6%8E%A8%E5%AF%BC%E6%96%B9%E6%B3%95.pdf) - 6、[面向Python程序源代码的分析与编译优化研究_范浩杰](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/05%E7%B1%BB%E5%9E%8B%E7%B3%BB%E7%BB%9F%E4%B8%8E%E9%9D%99%E6%80%81%E5%88%86%E6%9E%90/%E9%9D%A2%E5%90%91Python%E7%A8%8B%E5%BA%8F%E6%BA%90%E4%BB%A3%E7%A0%81%E7%9A%84%E5%88%86%E6%9E%90%E4%B8%8E%E7%BC%96%E8%AF%91%E4%BC%98%E5%8C%96%E7%A0%94%E7%A9%B6.pdf) - 7、[编译器ZOMI_12前端优化概述](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/%E7%BC%96%E8%AF%91%E5%99%A8ZOMI/11AI%E7%BC%96%E8%AF%91%E5%99%A8%E7%9A%84%E6%8C%91%E6%88%98%E4%B8%8E%E6%80%9D%E8%80%83.pdf) - 8、[编译器ZOMI_13计算图层IR](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/%E7%BC%96%E8%AF%91%E5%99%A8ZOMI/13%E8%AE%A1%E7%AE%97%E5%9B%BE%E5%B1%82IR.pdf) - 9、[编译器ZOMI_14算子融合策略](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/%E7%BC%96%E8%AF%91%E5%99%A8ZOMI/14%E7%AE%97%E5%AD%90%E8%9E%8D%E5%90%88%E7%AD%96%E7%95%A5.pdf) - 10、[编译器ZOMI_15布局转换原理](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/%E7%BC%96%E8%AF%91%E5%99%A8ZOMI/15%E5%B8%83%E5%B1%80%E8%BD%AC%E6%8D%A2%E5%8E%9F%E7%90%86.pdf) - 11、[编译器ZOMI_16布局转换算法](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/%E7%BC%96%E8%AF%91%E5%99%A8ZOMI/16%E5%B8%83%E5%B1%80%E8%BD%AC%E6%8D%A2%E7%AE%97%E6%B3%95.pdf) - 12、[编译器ZOMI_17内存分配算法](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/%E7%BC%96%E8%AF%91%E5%99%A8ZOMI/17%E5%86%85%E5%AD%98%E5%88%86%E9%85%8D%E7%AE%97%E6%B3%95.pdf) - 13、[编译器ZOMI_18常量折叠](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/%E7%BC%96%E8%AF%91%E5%99%A8ZOMI/18%E5%B8%B8%E9%87%8F%E6%8A%98%E5%8F%A0.pdf) - 14、[编译器ZOMI_19公共子表达式消除](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/%E7%BC%96%E8%AF%91%E5%99%A8ZOMI/19%E5%85%AC%E5%85%B1%E8%A1%A8%E8%BE%BE%E5%BC%8F%E6%B6%88%E9%99%A4.pdf) - 15、[编译器ZOMI_20死代码消除](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/%E7%BC%96%E8%AF%91%E5%99%A8ZOMI/20%E6%AD%BB%E4%BB%A3%E7%A0%81%E6%B6%88%E9%99%A4.pdf) - 16、[编译器ZOMI_21代数化简原理](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/%E7%BC%96%E8%AF%91%E5%99%A8ZOMI/21%E4%BB%A3%E6%95%B0%E5%8C%96%E7%AE%80%E5%8E%9F%E7%90%86.pdf) - 17、[编译器ZOMI_22前端优化总结](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/%E7%BC%96%E8%AF%91%E5%99%A8ZOMI/22%E5%89%8D%E7%AB%AF%E4%BC%98%E5%8C%96%E6%80%BB%E7%BB%93.pdf) - 18、[编译器ZOMI_29PyTorch2.0特性串讲](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/%E7%BC%96%E8%AF%91%E5%99%A8ZOMI/29PyTorch2.0%E7%89%B9%E6%80%A7%E4%B8%B2%E8%AE%B2.pdf) - 19、[编译器ZOMI_30TorchScript静态图串讲](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/%E7%BC%96%E8%AF%91%E5%99%A8ZOMI/30TorchScript%E9%9D%99%E6%80%81%E5%9B%BE%E5%B0%9D%E8%AF%95.pdf) - 20、[编译器ZOMI_31Torch FX与Lazytensor特性](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/%E7%BC%96%E8%AF%91%E5%99%A8ZOMI/31Torch%20FX%E4%B8%8ELazyTensor%E7%89%B9%E6%80%A7.pdf) - 21、[编译器ZOMI_32TorchDynamo](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/%E7%BC%96%E8%AF%91%E5%99%A8ZOMI/32TorchDynamo.pdf) - 22、[编译器ZOMI_33AOTAutograd原理](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/%E7%BC%96%E8%AF%91%E5%99%A8ZOMI/33AOTAutograd%E5%8E%9F%E7%90%86.pdf) - 23、[编译器ZOMI_34Dispatch机制](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/%E7%BC%96%E8%AF%91%E5%99%A8ZOMI/34Dispatch%E6%9C%BA%E5%88%B6.pdf) - 24、[计算图ZOMI_01计算图基础介绍](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/06%E8%AE%A1%E7%AE%97%E5%9B%BE%E7%94%9F%E6%88%90/%E8%AE%A1%E7%AE%97%E5%9B%BEZOMI/01%E8%AE%A1%E7%AE%97%E5%9B%BE%E5%9F%BA%E7%A1%80%E4%BB%8B%E7%BB%8D%20.pdf) - 25、[计算图ZOMI_02计算图（数据流图）](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/06%E8%AE%A1%E7%AE%97%E5%9B%BE%E7%94%9F%E6%88%90/%E8%AE%A1%E7%AE%97%E5%9B%BEZOMI/02%E8%AE%A1%E7%AE%97%E5%9B%BE(%E6%95%B0%E6%8D%AE%E6%B5%81%E5%9B%BE).pdf) - 24、[计算图ZOMI_03计算图和自动微分](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/06%E8%AE%A1%E7%AE%97%E5%9B%BE%E7%94%9F%E6%88%90/%E8%AE%A1%E7%AE%97%E5%9B%BEZOMI/03%E8%AE%A1%E7%AE%97%E5%9B%BE%E5%92%8C%E8%87%AA%E5%8A%A8%E5%BE%AE%E5%88%86.pdf) - 25、[计算图ZOMI_04图的调度与执行](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/06%E8%AE%A1%E7%AE%97%E5%9B%BE%E7%94%9F%E6%88%90/%E8%AE%A1%E7%AE%97%E5%9B%BEZOMI/04%E5%9B%BE%E7%9A%84%E8%B0%83%E5%BA%A6%E4%B8%8E%E6%89%A7%E8%A1%8C.pdf) - 24、[计算图ZOMI_05图与控制流](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/06%E8%AE%A1%E7%AE%97%E5%9B%BE%E7%94%9F%E6%88%90/%E8%AE%A1%E7%AE%97%E5%9B%BEZOMI/05%E5%9B%BE%E4%B8%8E%E6%8E%A7%E5%88%B6%E6%B5%81.pdf) - 25、[计算图ZOMI_06计算图的挑战与未来](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/06%E8%AE%A1%E7%AE%97%E5%9B%BE%E7%94%9F%E6%88%90/%E8%AE%A1%E7%AE%97%E5%9B%BEZOMI/06%E8%AE%A1%E7%AE%97%E5%9B%BE%E7%9A%84%E6%8C%91%E6%88%98%E4%B8%8E%E6%9C%AA%E6%9D%A5.pdf) - 26、[System_for_AI Microsoft_09计算图的编译与优化](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/System_for_AI%20Microsoft/09%E8%AE%A1%E7%AE%97%E5%9B%BE%E7%9A%84%E7%BC%96%E8%AF%91%E4%B8%8E%E4%BC%98%E5%8C%96.pdf) #### 07中间表示 - 1、[MLIR入门-张洪宾](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/07%E4%B8%AD%E9%97%B4%E8%A1%A8%E7%A4%BA/MLIR%E5%85%A5%E9%97%A8-%E5%BC%A0%E6%B4%AA%E5%AE%BE.docx) - 2、[面向机器学习系统的张量中间表示_庄陈敏](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/07%E4%B8%AD%E9%97%B4%E8%A1%A8%E7%A4%BA/%E5%8F%82%E8%80%83%E8%B5%84%E6%96%99/A%20tensor%20inte...rning%20systems_%E5%BA%84%E6%AF%85%E6%95%8F.pdf) - 3、[An IR for ML Pipelines](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/07%E4%B8%AD%E9%97%B4%E8%A1%A8%E7%A4%BA/%E5%8F%82%E8%80%83%E8%B5%84%E6%96%99/An%20IR%20for%20ML%20Pipelines.pdf) - 4、[IR2013](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/07%E4%B8%AD%E9%97%B4%E8%A1%A8%E7%A4%BA/%E5%8F%82%E8%80%83%E8%B5%84%E6%96%99/IR2013.pdf) - 5、[Relay: A new ir for machine learning frameworks by Roesch J, Lyubomirsky S, Weber L, et al.2018](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/07%E4%B8%AD%E9%97%B4%E8%A1%A8%E7%A4%BA/%E5%8F%82%E8%80%83%E8%B5%84%E6%96%99/TVM_RelayIR.pdf) - 6、[TensorIR: An Abstraction for Automatic Tensorized Program Optimization by Siyuan Feng, Bohan Hou et al., arXiv 2022](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/07%E4%B8%AD%E9%97%B4%E8%A1%A8%E7%A4%BA/%E5%8F%82%E8%80%83%E8%B5%84%E6%96%99/TVM_TIR.pdf) - 7、[The Deep Learning Compiler: A Comprehensive Survey by Mingzhen Li et al., TPDS 2020](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/07%E4%B8%AD%E9%97%B4%E8%A1%A8%E7%A4%BA/%E5%8F%82%E8%80%83%E8%B5%84%E6%96%99/The%20Deep%20Learning%20Compiler%20A%20Comprehensive%20Survey.pdf) - 8、[一种深度学习编译器的高层中间表示转换方法及相关装置](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/07%E4%B8%AD%E9%97%B4%E8%A1%A8%E7%A4%BA/%E5%8F%82%E8%80%83%E8%B5%84%E6%96%99/%E4%B8%80%E7%A7%8D%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E5%99%A8%E7%9A%84%E9%AB%98%E5%B1%82%E4%B8%AD%E9%97%B4%E8%A1%A8%E7%A4%BA%E8%BD%AC%E6%8D%A2%E6%96%B9%E6%B3%95%E5%8F%8A%E7%9B%B8%E5%85%B3%E8%A3%85%E7%BD%AE.pdf) - 9、[一种面向神经网络模型计算的中间表示方法和装置](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/07%E4%B8%AD%E9%97%B4%E8%A1%A8%E7%A4%BA/%E5%8F%82%E8%80%83%E8%B5%84%E6%96%99/%E4%B8%80%E7%A7%8D%E9%9D%A2%E5%90%91%E7%A5%9E%E7%BB%8F%E7%BD%91%E7%BB%9C%E6%A8%A1%E5%9E%8B%E8%AE%A1%E7%AE%97%E7%9A%84%E4%B8%AD%E9%97%B4%E8%A1%A8%E7%A4%BA%E6%96%B9%E6%B3%95%E5%92%8C%E8%A3%85%E7%BD%AE.pdf) - 10、[在编译器中构建基于图的中间表示的方法](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/07%E4%B8%AD%E9%97%B4%E8%A1%A8%E7%A4%BA/%E5%8F%82%E8%80%83%E8%B5%84%E6%96%99/%E5%9C%A8%E7%BC%96%E8%AF%91%E5%99%A8%E4%B8%AD%E6%9E%84%E5%BB%BA%E5%9F%BA%E4%BA%8E%E5%9B%BE%E7%9A%84%E4%B8%AD%E9%97%B4%E8%A1%A8%E7%A4%BA%E7%9A%84%E6%96%B9%E6%B3%95.pdf) - 11、[用于根据TensorFlow图构建编译器中间表示的方法和系统](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/07%E4%B8%AD%E9%97%B4%E8%A1%A8%E7%A4%BA/%E5%8F%82%E8%80%83%E8%B5%84%E6%96%99/%E7%94%A8%E4%BA%8E%E6%A0%B9%E6%8D%AETensorFlow%E5%9B%BE%E6%9E%84%E5%BB%BA%E7%BC%96%E8%AF%91%E5%99%A8%E4%B8%AD%E9%97%B4%E8%A1%A8%E7%A4%BA%E7%9A%84%E6%96%B9%E6%B3%95%E5%92%8C%E7%B3%BB%E7%BB%9F.pdf) - 12、[主流IR相关帖子](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/07%E4%B8%AD%E9%97%B4%E8%A1%A8%E7%A4%BA/%E5%8F%82%E8%80%83%E8%B5%84%E6%96%99/%E4%B8%BB%E6%B5%81IR%E7%9B%B8%E5%85%B3%E5%B8%96%E5%AD%90) #### 08自动微分 - 1、[Automatic Differentiation in Machine Learning: a Survey](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/08%E8%87%AA%E5%8A%A8%E5%BE%AE%E5%88%86/Automatic%20Differentiation%20in%20Machine%20Learning%20a%20Survey.pdf) - 2、[A Brief Introduction to Automatic Differentiation for Machine Learning](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/08%E8%87%AA%E5%8A%A8%E5%BE%AE%E5%88%86/A%20Brief%20Introduction%20to%20Automatic%20Differentiation%20for%20Machine%20Learning.pdf) - 3、[automatic differentiation in ml where we are and where we should begoing](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/08%E8%87%AA%E5%8A%A8%E5%BE%AE%E5%88%86/automatic-differentiation-in-ml-where-we-are-and-where-we-should-be-going.pdf) - 4、[两种计算偏微分方程数值解的神经网络方法_韩祖良](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/08%E8%87%AA%E5%8A%A8%E5%BE%AE%E5%88%86/%E4%B8%A4%E7%A7%8D%E8%AE%A1%E7%AE%97%E5%81%8F%E5%BE%AE%E5%88%86%E6%96%B9%E7%A8%8B%E6%95%B0%E5%80%BC%E8%A7%A3%E7%9A%84%E7%A5%9E%E7%BB%8F%E7%BD%91%E7%BB%9C%E6%96%B9%E6%B3%95_%E9%9F%A9%E7%A5%96%E8%89%AF.pdf) - 5、[微分万物：深度学习的启示](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/08%E8%87%AA%E5%8A%A8%E5%BE%AE%E5%88%86/%E5%BE%AE%E5%88%86%E4%B8%87%E7%89%A9%EF%BC%9A%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%9A%84%E5%90%AF%E7%A4%BA_%E7%8E%8B%E7%A3%8A.pdf) - 6、[Some highlights on Source-to-Source Adjoint AD](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/08%E8%87%AA%E5%8A%A8%E5%BE%AE%E5%88%86/some_highlights_on_source_to_s.pdf) - 7、[自动微分ZOMI_01自动微分概述](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/08%E8%87%AA%E5%8A%A8%E5%BE%AE%E5%88%86/%E8%87%AA%E5%8A%A8%E5%BE%AE%E5%88%86ZOMI/01%E8%87%AA%E5%8A%A8%E5%BE%AE%E5%88%86%E6%A6%82%E8%BF%B0.pdf) - 8、[自动微分ZOMI_02微分基本概念](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/08%E8%87%AA%E5%8A%A8%E5%BE%AE%E5%88%86/%E8%87%AA%E5%8A%A8%E5%BE%AE%E5%88%86ZOMI/02%E5%BE%AE%E5%88%86%E5%9F%BA%E6%9C%AC%E6%A6%82%E5%BF%B5.pdf) - 9、[自动微分ZOMI_03自动微分模式](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/08%E8%87%AA%E5%8A%A8%E5%BE%AE%E5%88%86/%E8%87%AA%E5%8A%A8%E5%BE%AE%E5%88%86ZOMI/03%E8%87%AA%E5%8A%A8%E5%BE%AE%E5%88%86%E6%A8%A1%E5%BC%8F.pdf) - 10、[自动微分ZOMI_04具体实现方式](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/08%E8%87%AA%E5%8A%A8%E5%BE%AE%E5%88%86/%E8%87%AA%E5%8A%A8%E5%BE%AE%E5%88%86ZOMI/04%E5%85%B7%E4%BD%93%E5%AE%9E%E7%8E%B0%E6%96%B9%E5%BC%8F.pdf) - 11、自动微分ZOMI_05前向模式[PDF](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/08%E8%87%AA%E5%8A%A8%E5%BE%AE%E5%88%86/%E8%87%AA%E5%8A%A8%E5%BE%AE%E5%88%86ZOMI/05%E5%89%8D%E5%90%91%E6%A8%A1%E5%BC%8F.pdf) [ipynb](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/08%E8%87%AA%E5%8A%A8%E5%BE%AE%E5%88%86/%E8%87%AA%E5%8A%A8%E5%BE%AE%E5%88%86ZOMI/05.forward_mode.ipynb) - 12、自动微分ZOMI_06反向模式[PDF](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/08%E8%87%AA%E5%8A%A8%E5%BE%AE%E5%88%86/%E8%87%AA%E5%8A%A8%E5%BE%AE%E5%88%86ZOMI/06%E5%8F%8D%E5%90%91%E6%A8%A1%E5%BC%8F.pdf) [ipynb](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/08%E8%87%AA%E5%8A%A8%E5%BE%AE%E5%88%86/%E8%87%AA%E5%8A%A8%E5%BE%AE%E5%88%86ZOMI/06.reversed_mode.ipynb) - 13、[自动微分ZOMI_07自动微分未来挑战](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/08%E8%87%AA%E5%8A%A8%E5%BE%AE%E5%88%86/%E8%87%AA%E5%8A%A8%E5%BE%AE%E5%88%86ZOMI/07%E8%87%AA%E5%8A%A8%E5%BE%AE%E5%88%86%E6%9C%AA%E6%9D%A5%E6%8C%91%E6%88%98.pdf) - 14、[计算图ZOMI_03计算图和自动微分](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/06%E8%AE%A1%E7%AE%97%E5%9B%BE%E7%94%9F%E6%88%90/%E8%AE%A1%E7%AE%97%E5%9B%BEZOMI/03%E8%AE%A1%E7%AE%97%E5%9B%BE%E5%92%8C%E8%87%AA%E5%8A%A8%E5%BE%AE%E5%88%86.pdf) #### 09计算图优化 - 1、[AStitch: Enabling a New Multi-dimensional Optimization Space for Memory-Intensive ML Training and Inference on Modern SIMT Architectures by Zhen Zheng et al., ASPLOS 2022]( https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/09%E8%AE%A1%E7%AE%97%E5%9B%BE%E4%BC%98%E5%8C%96/Astitch.pdf) - 2、[Apollo: Automatic Partition-based Operator Fusion through Layer by Layer Optimization by Jie Zhao et al., MLSys 2022](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/09%E8%AE%A1%E7%AE%97%E5%9B%BE%E4%BC%98%E5%8C%96/MLSys-2022-apollo-automatic-partition-based-operator-fusion-through-layer-by-layer-optimization-Paper.pdf) - 3、[Cortex: A Compiler for Recursive Deep Learning Models by Pratik Fegade et al., MLSys 2021](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/09%E8%AE%A1%E7%AE%97%E5%9B%BE%E4%BC%98%E5%8C%96/MLSys21-cortex.pdf) - 4、[DISC: A Dynamic Shape Compiler for Machine Learning Workloads by Kai Zhu et al., EuroMLSys 2021](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/09%E8%AE%A1%E7%AE%97%E5%9B%BE%E4%BC%98%E5%8C%96/DISC.pdf) - 5、[GSO_基于图神经网络的深...学习计算图子图替换优化框架_苗旭鹏](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/09%E8%AE%A1%E7%AE%97%E5%9B%BE%E4%BC%98%E5%8C%96/GSO_%E5%9F%BA%E4%BA%8E%E5%9B%BE%E7%A5%9E%E7%BB%8F%E7%BD%91%E7%BB%9C%E7%9A%84%E6%B7%B1...%E5%AD%A6%E4%B9%A0%E8%AE%A1%E7%AE%97%E5%9B%BE%E5%AD%90%E5%9B%BE%E6%9B%BF%E6%8D%A2%E4%BC%98%E5%8C%96%E6%A1%86%E6%9E%B6_%E8%8B%97%E6%97%AD%E9%B9%8F.pdf) - 6、[Equality Saturation for Tensor Graph Superoptimization by Yichen Yang et al., MLSys 2021](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/09%E8%AE%A1%E7%AE%97%E5%9B%BE%E4%BC%98%E5%8C%96/Equality%20Saturation%20for%20Tensor%20Graph%20Superoptimization.pdf) - 7、[FusionStitching: Boosting Memory IntensiveComputations for Deep Learning Workloads by Zhen Zheng et al., arXiv 2020](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/09%E8%AE%A1%E7%AE%97%E5%9B%BE%E4%BC%98%E5%8C%96/FusionStitching.pdf) - 8、[Nimble: Lightweight and Parallel GPU Task Scheduling for Deep Learning by Woosuk Kwon et al., Neurips 2020](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/09%E8%AE%A1%E7%AE%97%E5%9B%BE%E4%BC%98%E5%8C%96/NeurIPS-2020-nimble-lightweight-and-parallel-gpu-task-scheduling-for-deep-learning-Paper.pdf) - 9、[TensorFlow 设计白皮书](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/09%E8%AE%A1%E7%AE%97%E5%9B%BE%E4%BC%98%E5%8C%96/Tensorflow%E8%AE%BE%E8%AE%A1%E7%99%BD%E7%9A%AE%E4%B9%A6.pdf) - 10、[MindSpore技术白皮书](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/09%E8%AE%A1%E7%AE%97%E5%9B%BE%E4%BC%98%E5%8C%96/MindSpore_white_paperV1.1.pdf) - 11、[Rammer: Enabling Holistic Deep Learning Compiler Optimizations with rTasks by Lingxiao Ma et al., OSDI 2020](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/09%E8%AE%A1%E7%AE%97%E5%9B%BE%E4%BC%98%E5%8C%96/osdi20-Rammer.pdf) - 12、[Relay: A High-Level Compiler for Deep Learning by Jared Roesch et al., arXiv 2019](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/09%E8%AE%A1%E7%AE%97%E5%9B%BE%E4%BC%98%E5%8C%96/Relay.pdf) - 13、[The CoRa Tensor Compiler: Compilation for Ragged Tensors with Minimal Padding by Pratik Fegade et al., MLSys 2022](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/09%E8%AE%A1%E7%AE%97%E5%9B%BE%E4%BC%98%E5%8C%96/MLSys-2022-the-cora-tensor-compiler-compilation-for-ragged-tensors-with-minimal-padding-Paper.pdf) - 14、[Roller: Fast and Efficient Tensor Compilation for Deep Learning by Hongyu Zhu et al., OSDI 2022](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/09%E8%AE%A1%E7%AE%97%E5%9B%BE%E4%BC%98%E5%8C%96/osdi22-Roller.pdf) - 15、[TensorFlow Eager论文](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/09%E8%AE%A1%E7%AE%97%E5%9B%BE%E4%BC%98%E5%8C%96/TensorFlow%20Eager%E8%AE%BA%E6%96%87.pdf) - 16、[TensorFlow 控制流](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/09%E8%AE%A1%E7%AE%97%E5%9B%BE%E4%BC%98%E5%8C%96/TensorFlow%E6%8E%A7%E5%88%B6%E6%B5%81.pdf) #### 10算子生成与优化 - 1、[TVM: An Automated End-to-End Optimizing Compiler for Deep Learning by Tianqi Chen et al., OSDI 2018](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/10%E7%AE%97%E5%AD%90%E7%94%9F%E6%88%90%E5%8F%8A%E4%BC%98%E5%8C%96/1.TVM2017.pdf) - 2、[Relay: A new ir for machine learning frameworks by Roesch J, Lyubomirsky S, Weber L, et al.2018](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/10%E7%AE%97%E5%AD%90%E7%94%9F%E6%88%90%E5%8F%8A%E4%BC%98%E5%8C%96/2.Relay18.pdf) - 3、[Ansor: Generating High-Performance Tensor Programs for Deep Learning by Lianmin Zheng et al., OSDI 2020](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/10%E7%AE%97%E5%AD%90%E7%94%9F%E6%88%90%E5%8F%8A%E4%BC%98%E5%8C%96/Ansor.pdf) - 4、[GSO_基于图神经网络的深度学习计算图子图替换优化框架_苗旭鹏](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/10%E7%AE%97%E5%AD%90%E7%94%9F%E6%88%90%E5%8F%8A%E4%BC%98%E5%8C%96/GSO_%E5%9F%BA%E4%BA%8E%E5%9B%BE%E7%A5%9E%E7%BB%8F%E7%BD%91%E7%BB%9C%E7%9A%84%E6%B7%B1...%E5%AD%A6%E4%B9%A0%E8%AE%A1%E7%AE%97%E5%9B%BE%E5%AD%90%E5%9B%BE%E6%9B%BF%E6%8D%A2%E4%BC%98%E5%8C%96%E6%A1%86%E6%9E%B6_%E8%8B%97%E6%97%AD%E9%B9%8F.pdf) - 5、[一种用于深度学习编译器中探索优化空间的加速方法_潘秋红](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/10%E7%AE%97%E5%AD%90%E7%94%9F%E6%88%90%E5%8F%8A%E4%BC%98%E5%8C%96/%E4%B8%80%E7%A7%8D%E7%94%A8%E4%BA%8E%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E5%99%A8%E4%B8%AD%E6%8E%A2%E7%B4%A2%E4%BC%98%E5%8C%96%E7%A9%BA%E9%97%B4%E7%9A%84%E5%8A%A0%E9%80%9F%E6%96%B9%E6%B3%95_%E6%BD%98%E7%A7%8B%E7%BA%A2.pdf) - 6、[一种面向深度学习编译器的高效算子优化方法_孟晓](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/10%E7%AE%97%E5%AD%90%E7%94%9F%E6%88%90%E5%8F%8A%E4%BC%98%E5%8C%96/%E4%B8%80%E7%A7%8D%E9%9D%A2%E5%90%91%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E5%99%A8%E7%9A%84%E9%AB%98%E6%95%88%E7%AE%97%E5%AD%90%E4%BC%98%E5%8C%96%E6%96%B9%E6%B3%95_%E5%AD%9F%E6%99%93.pdf) - 7、[基于ARM处理器的深度学习优化技术研究_罗纪杰](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/10%E7%AE%97%E5%AD%90%E7%94%9F%E6%88%90%E5%8F%8A%E4%BC%98%E5%8C%96/%E5%9F%BA%E4%BA%8EARM%E5%A4%84%E7%90%86%E5%99%A8%E7%9A%84%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E4%BC%98%E5%8C%96%E6%8A%80%E6%9C%AF%E7%A0%94%E7%A9%B6_%E7%BD%97%E7%BA%AA%E6%9D%B0.pdf) - 8、[面向航天异构平台的深度学习编译器加速技术优化_刘功晗](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/10%E7%AE%97%E5%AD%90%E7%94%9F%E6%88%90%E5%8F%8A%E4%BC%98%E5%8C%96/%E9%9D%A2%E5%90%91%E8%88%AA%E5%A4%A9%E5%BC%82%E6%9E%84%E5%B9%B3%E5%8F%B0%E7%9A%84%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E5%99%A8%E5%8A%A0%E9%80%9F%E6%8A%80%E6%9C%AF%E4%BC%98%E5%8C%96_%E5%88%98%E5%8A%9F%E6%99%97.pdf) - 9、[The Deep Learning Compiler A Comprehensive Survey](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/10%E7%AE%97%E5%AD%90%E7%94%9F%E6%88%90%E5%8F%8A%E4%BC%98%E5%8C%96/%E7%BB%BC%E8%BF%B0/The%20Deep%20Learning%20Compiler%20A%20Comprehensive%20Survey.pdf) - 10、[xla_A LEARNED PERFORMANCE MODEL FOR TENSOR PROCESSING UNITS.pdf](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/10%E7%AE%97%E5%AD%90%E7%94%9F%E6%88%90%E5%8F%8A%E4%BC%98%E5%8C%96/xla/A%20LEARNED%20PERFORMANCE%20MODEL%20FOR%20TENSOR%20PROCESSING%20UNITS.pdf) - 11、[xla_Learned TPU Cost Model for XLA Tensor Programs.pdf](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/10%E7%AE%97%E5%AD%90%E7%94%9F%E6%88%90%E5%8F%8A%E4%BC%98%E5%8C%96/xla/Learned%20TPU%20Cost%20Model%20for%20XLA%20Tensor%20Programs.pdf) - 12、[Halide A Language and Compiler for Optimizing Parallelism.pdf](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/10%E7%AE%97%E5%AD%90%E7%94%9F%E6%88%90%E5%8F%8A%E4%BC%98%E5%8C%96/Halide%20A%20Language%20and%20Compiler%20for%20Optimizing%20Parallelism.pdf) - 13、[Tensor Comprehensions.pdf](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/10%E7%AE%97%E5%AD%90%E7%94%9F%E6%88%90%E5%8F%8A%E4%BC%98%E5%8C%96/Tensor%20Comprehensions.pdf) - 14、[GLOW.pdf](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/10%E7%AE%97%E5%AD%90%E7%94%9F%E6%88%90%E5%8F%8A%E4%BC%98%E5%8C%96/GLOW.pdf) #### 11内存优化 - 1、[Layer-Centric Memory Reuse and Data Migration for Extreme-Scale Deep Learning on Many-Core Architectures](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/11%E5%86%85%E5%AD%98%E4%BC%98%E5%8C%96/Layer-Centric%20Memory%20Reuse%20and%20Data%20Migration%20for%20%20Extreme-Scale%20Deep%20Learning%20on%20Many-Core%20Architectures.pdf) - 2、[Training Deep Nets with Sublinear Memory Cost](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/11%E5%86%85%E5%AD%98%E4%BC%98%E5%8C%96/Training%20Deep%20Nets%20with%20Sublinear%20Memory%20Cost.pdf) - 3、[SuperNeurons: Dynamic GPU Memory Management for Training Deep Neural Network](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/11%E5%86%85%E5%AD%98%E4%BC%98%E5%8C%96/Dynamic%20GPU%20Memory%20Management%20for%20Training%20Deep%20Neural%20Network.pdf) - 4、[编译器ZOMI_23AI编译器后端优化介绍](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/%E7%BC%96%E8%AF%91%E5%99%A8ZOMI/23AI%E7%BC%96%E8%AF%91%E5%99%A8%E5%90%8E%E7%AB%AF%E4%BC%98%E5%8C%96%E4%BB%8B%E7%BB%8D.pdf) - 5、[编译器ZOMI_24算子的计算与调度](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/%E7%BC%96%E8%AF%91%E5%99%A8ZOMI/24%E7%AE%97%E5%AD%90%E7%9A%84%E8%AE%A1%E7%AE%97%E4%B8%8E%E8%B0%83%E5%BA%A6.pdf) - 6、[编译器ZOMI_25算子优化手工方式](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/%E7%BC%96%E8%AF%91%E5%99%A8ZOMI/25%E7%AE%97%E5%AD%90%E4%BC%98%E5%8C%96%E6%89%8B%E5%B7%A5%E6%96%B9%E5%BC%8F.pdf) - 6、[编译器ZOMI_26算子循环优化](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/%E7%BC%96%E8%AF%91%E5%99%A8ZOMI/26%E7%AE%97%E5%AD%90%E5%BE%AA%E7%8E%AF%E4%BC%98%E5%8C%96.pdf) - 7、[编译器ZOMI_27指令和内存优化](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/%E7%BC%96%E8%AF%91%E5%99%A8ZOMI/27%E6%8C%87%E4%BB%A4%E5%92%8C%E5%86%85%E5%AD%98%E4%BC%98%E5%8C%96.pdf) - 8、[编译器ZOMI_28Auto-Tuning原理](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/%E7%BC%96%E8%AF%91%E5%99%A8ZOMI/28Auto-Tuning%E5%8E%9F%E7%90%86.pdf) #### 12代码生成和执行 - 1、[swTVM Exploring the Automated Compilation for Deep Learning on Sunway Architecture.pdf](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/12%E4%BB%A3%E7%A0%81%E7%94%9F%E6%88%90%E5%92%8C%E6%89%A7%E8%A1%8C/swTVM%20Exploring%20the%20Automated%20Compilation%20for%20Deep%20Learning%20on%20Sunway%20Architecture.pdf) - 2、[一种基于TVM编译器的异构平台的部署方法及装置_吴金进.pdf](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/12%E4%BB%A3%E7%A0%81%E7%94%9F%E6%88%90%E5%92%8C%E6%89%A7%E8%A1%8C/%E4%B8%80%E7%A7%8D%E5%9F%BA%E4%BA%8ETVM%E7%BC%96%E8%AF%91%E5%99%A8%E7%9A%84%E5%BC%82%E6%9E%84%E5%B9%B3%E5%8F%B0%E7%9A%84%E9%83%A8%E7%BD%B2%E6%96%B9%E6%B3%95%E5%8F%8A%E8%A3%85%E7%BD%AE_%E5%90%B4%E9%87%91%E8%BF%9B.pdf) - 3、[基于申威处理器的深度学习算子自动优化系统及方法_杨广文.pdf](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/12%E4%BB%A3%E7%A0%81%E7%94%9F%E6%88%90%E5%92%8C%E6%89%A7%E8%A1%8C/%E5%9F%BA%E4%BA%8E%E7%94%B3%E5%A8%81%E5%A4%84%E7%90%86%E5%99%A8%E7%9A%84%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%AE%97%E5%AD%90%E8%87%AA%E5%8A%A8%E4%BC%98%E5%8C%96%E7%B3%BB%E7%BB%9F%E5%8F%8A%E6%96%B9%E6%B3%95_%E6%9D%A8%E5%B9%BF%E6%96%87.pdf) #### 13分布式训练 - 1、[分布式深度学习训练网络综述](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E5%88%86%E5%B8%83%E5%BC%8F%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E8%AE%AD%E7%BB%83%E7%BD%91%E7%BB%9C%E7%BB%BC%E8%BF%B0_%E6%9C%B1%E6%B3%93%E7%9D%BF.pdf) - 2、[分布式机器学习算法收敛敏感性优化技术研究_范禹辰](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83_%E5%88%86%E5%B8%83%E5%BC%8F%E6%9C%BA%E5%99%A8%E5%AD%A6%E4%B9%A0%E7%AE%97%E6%B3%95%E6%94%B6%E6%95%9B%E6%95%8F%E6%84%9F%E6%80%A7%E4%BC%98%E5%8C%96%E6%8A%80%E6%9C%AF%E7%A0%94%E7%A9%B6_%E8%8C%83%E7%A6%B9%E8%BE%B0.pdf) - 3、[分布式机器学习集群系统性能优化_和新树](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83_%E5%88%86%E5%B8%83%E5%BC%8F%E6%9C%BA%E5%99%A8%E5%AD%A6%E4%B9%A0%E9%9B%86%E7%BE%A4%E7%B3%BB%E7%BB%9F%E6%80%A7%E8%83%BD%E4%BC%98%E5%8C%96_%E5%92%8C%E6%96%B0%E6%A0%91.pdf) - 4、[基于分布式机器学习的车辆定位方法研究_张恩龙](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83_%E5%9F%BA%E4%BA%8E%E5%88%86%E5%B8%83%E5%BC%8F%E6%9C%BA%E5%99%A8%E5%AD%A6%E4%B9%A0%E7%9A%84%E8%BD%A6%E8%BE%86%E5%AE%9A%E4%BD%8D%E6%96%B9%E6%B3%95%E7%A0%94%E7%A9%B6_%E5%BC%A0%E6%81%A9%E9%BE%99.pdf) - 5、[基于数据和模型混合传输的分布式机器学习框架_严佳媚](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83_%E5%9F%BA%E4%BA%8E%E6%95%B0%E6%8D%AE%E5%92%8C%E6%A8%A1%E5%9E%8B%E6%B7%B7%E5%90%88%E4%BC%A0%E8%BE%93%E7%9A%84%E5%88%86%E5%B8%83%E5%BC%8F%E6%9C%BA%E5%99%A8%E5%AD%A6%E4%B9%A0%E6%A1%86%E6%9E%B6_%E4%B8%A5%E4%BD%B3%E5%AA%9A.pdf) - 6、[分布式训练_面向分布式机器学习框架的通信优化技术研究_阳瑞](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83_%E9%9D%A2%E5%90%91%E5%88%86%E5%B8%83%E5%BC%8F%E6%9C%BA%E5%99%A8%E5%AD%A6%E4%B9%A0%E6%A1%86%E6%9E%B6%E7%9A%84%E9%80%9A%E4%BF%A1%E4%BC%98%E5%8C%96%E6%8A%80%E6%9C%AF%E7%A0%94%E7%A9%B6_%E9%98%B3%E7%91%9E.pdf) - 7、[深度学习分布式训练并行策略的研究_万知雨](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83%E5%B9%B6%E8%A1%8C%E7%AD%96%E7%95%A5%E7%9A%84%E7%A0%94%E7%A9%B6_%E4%B8%87%E7%9F%A5%E9%9B%A8.pdf) - 8、[深度学习并行分布式训练机制研究_姚琼杰](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E5%B9%B6%E8%A1%8C%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83%E6%9C%BA%E5%88%B6%E7%A0%94%E7%A9%B6_%E5%A7%9A%E7%90%BC%E6%9D%B0.pdf) - 9、[深度学习网络分布式训练方案研究与性能优化_张泽超](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BD%91%E7%BB%9C%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83%E6%96%B9%E6%A1%88%E7%A0%94%E7%A9%B6%E4%B8%8E%E6%80%A7%E8%83%BD%E4%BC%98%E5%8C%96_%E5%BC%A0%E6%B3%BD%E8%B6%85.pdf) - 10、[深度神经网络并行化研究综述_朱虎明](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E6%B7%B1%E5%BA%A6%E7%A5%9E%E7%BB%8F%E7%BD%91%E7%BB%9C%E5%B9%B6%E8%A1%8C%E5%8C%96%E7%A0%94%E7%A9%B6%E7%BB%BC%E8%BF%B0_%E6%9C%B1%E8%99%8E%E6%98%8E.pdf) - 11、[针对深度学习模型的优化问题研究_郑书新](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E9%92%88%E5%AF%B9%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E6%A8%A1%E5%9E%8B%E7%9A%84%E4%BC%98%E5%8C%96%E9%97%AE%E9%A2%98%E7%A0%94%E7%A9%B6_%E9%83%91%E4%B9%A6%E6%96%B0.pdf) - 12、[自动并行_Exploring Hidden Dimensions in Parallelizing Convolutional Neural Networks](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E8%87%AA%E5%8A%A8%E5%B9%B6%E8%A1%8C/1%E3%80%81Exploring%20Hidden%20Dimensions%20in%20Parallelizing%20Convolutional%20Neural%20Networks.pdf) - 13、[自动并行_BEYOND DATA AND MODEL PARALLELISM FOR DEEP NEURAL NETWORKS](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E8%87%AA%E5%8A%A8%E5%B9%B6%E8%A1%8C/2%E3%80%81BEYOND%20DATA%20AND%20MODEL%20PARALLELISM%20FOR%20DEEP%20NEURAL%20NETWORKS.pdf) - 14、[自动并行_Device Placement Optimization with Reinforcement Learning](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E8%87%AA%E5%8A%A8%E5%B9%B6%E8%A1%8C/3%E3%80%81Device%20Placement%20Optimization%20with%20Reinforcement%20Learning.pdf) - 15、[自动并行_OneFlow Redesign the Distributed Deep Learning Framework from Scratch](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E8%87%AA%E5%8A%A8%E5%B9%B6%E8%A1%8C/4%E3%80%81OneFlow%20Redesign%20the%20Distributed%20Deep%20Learning%20Framework%20from%20Scratch.pdf) - 16、[自动并行_Automap Towards Ergonomic Automated Parallelism for ML Models](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E8%87%AA%E5%8A%A8%E5%B9%B6%E8%A1%8C/5%E3%80%81Automap%20Towards%20Ergonomic%20Automated%20Parallelism%20for%20ML%20Models.pdf) - 17、[自动并行_Auto-MAP A DQN Framework for Exploring Distributed Execution Plans for DNN Workloads](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E8%87%AA%E5%8A%A8%E5%B9%B6%E8%A1%8C/6%E3%80%81Auto-MAP%20A%20DQN%20Framework%20for%20Exploring%20Distributed%20Execution%20Plans%20for%20DNN%20Workloads.pdf) - 18、[自动并行_GShard Scaling Giant Models with Conditional Computation and Automatic Sharding](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E8%87%AA%E5%8A%A8%E5%B9%B6%E8%A1%8C/7%E3%80%81GShard%20Scaling%20Giant%20Models%20with%20Conditional%20Computation%20and%20Automatic%20Sharding.pdf) - 19、[自动并行_GSPMD General and Scalable Parallelization for ML Computation Graphs](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E8%87%AA%E5%8A%A8%E5%B9%B6%E8%A1%8C/8%E3%80%81GSPMD%20General%20and%20Scalable%20Parallelization%20for%20ML%20Computation%20Graphs..pdf) - 20、[自动并行_TensorOpt Exploring the Tradeoffs in Distributed DNN Training with Auto-Parallelism](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E8%87%AA%E5%8A%A8%E5%B9%B6%E8%A1%8C/9%E3%80%81TensorOpt%20Exploring%20the%20Tradeoffs%20in%20Distributed%20DNN%20Training%20with%20Auto-Parallelism.pdf) - 21、[自动并行_贾志豪1、2 论文翻译](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E8%87%AA%E5%8A%A8%E5%B9%B6%E8%A1%8C/%E8%B4%BE%E5%BF%97%E8%B1%AA1%E3%80%812%20%E8%AE%BA%E6%96%87%E7%BF%BB%E8%AF%91.docx) - 22、[自动并行_论文简介](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E8%87%AA%E5%8A%A8%E5%B9%B6%E8%A1%8C/%E8%AE%BA%E6%96%87%E7%AE%80%E4%BB%8B.docx) - 23、[分布式ZOMI_01分布式集群系统基本介绍](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E5%88%86%E5%B8%83%E5%BC%8FZOMI/01%E5%88%86%E5%B8%83%E5%BC%8F%E9%9B%86%E7%BE%A4%E7%B3%BB%E7%BB%9F%E5%9F%BA%E6%9C%AC%E4%BB%8B%E7%BB%8D.pdf) - 24、[分布式ZOMI_02分布式集群_AI集群架构](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E5%88%86%E5%B8%83%E5%BC%8FZOMI/02%E5%88%86%E5%B8%83%E5%BC%8F%E9%9B%86%E7%BE%A4_AI%E9%9B%86%E7%BE%A4%E6%9E%B6%E6%9E%84.pdf) - 25、[分布式ZOMI_03分布式集群_AI集群软硬件通信](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E5%88%86%E5%B8%83%E5%BC%8FZOMI/03%E5%88%86%E5%B8%83%E5%BC%8F%E9%9B%86%E7%BE%A4_AI%E9%9B%86%E7%BE%A4%E8%BD%AF%E7%A1%AC%E4%BB%B6%E9%80%9A%E4%BF%A1.pdf) - 26、[分布式ZOMI_04分布式集群_分布式通信原语](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E5%88%86%E5%B8%83%E5%BC%8FZOMI/04%E5%88%86%E5%B8%83%E5%BC%8F%E9%9B%86%E7%BE%A4_%E5%88%86%E5%B8%83%E5%BC%8F%E9%80%9A%E4%BF%A1%E5%8E%9F%E8%AF%AD.pdf) - 27、[分布式ZOMI_05分布式集群_框架分布式功能](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E5%88%86%E5%B8%83%E5%BC%8FZOMI/05%E5%88%86%E5%B8%83%E5%BC%8F%E9%9B%86%E7%BE%A4_%E6%A1%86%E6%9E%B6%E5%88%86%E5%B8%83%E5%BC%8F%E5%8A%9F%E8%83%BD.pdf) - 28、[分布式ZOMI_06分布式算法_大模型训练挑战](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E5%88%86%E5%B8%83%E5%BC%8FZOMI/06%E5%88%86%E5%B8%83%E5%BC%8F%E7%AE%97%E6%B3%95_%E5%A4%A7%E6%A8%A1%E5%9E%8B%E8%AE%AD%E7%BB%83%E6%8C%91%E6%88%98.pdf) - 29、[分布式ZOMI_07分布式算法_大模型算法结构](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E5%88%86%E5%B8%83%E5%BC%8FZOMI/07%E5%88%86%E5%B8%83%E5%BC%8F%E7%AE%97%E6%B3%95_%E5%A4%A7%E6%A8%A1%E5%9E%8B%E7%AE%97%E6%B3%95%E7%BB%93%E6%9E%84.pdf) - 30、[分布式ZOMI_08分布式算法_SOTA大模型算法](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E5%88%86%E5%B8%83%E5%BC%8FZOMI/08%E5%88%86%E5%B8%83%E5%BC%8F%E7%AE%97%E6%B3%95_SOTA%E5%A4%A7%E6%A8%A1%E5%9E%8B%E7%AE%97%E6%B3%95.pdf) [简介](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E5%88%86%E5%B8%83%E5%BC%8FZOMI/08%E5%A4%A7%E6%A8%A1%E5%9E%8B%E7%AE%97%E6%B3%95/01.introduction.pptx) [Bert&GPT3](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E5%88%86%E5%B8%83%E5%BC%8FZOMI/08%E5%A4%A7%E6%A8%A1%E5%9E%8B%E7%AE%97%E6%B3%95/02.Bert&GPT3.pptx) [T5](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E5%88%86%E5%B8%83%E5%BC%8FZOMI/08%E5%A4%A7%E6%A8%A1%E5%9E%8B%E7%AE%97%E6%B3%95/03.T5.pptx) [dataset](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E5%88%86%E5%B8%83%E5%BC%8FZOMI/08%E5%A4%A7%E6%A8%A1%E5%9E%8B%E7%AE%97%E6%B3%95/dataset.pptx) [template](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E5%88%86%E5%B8%83%E5%BC%8FZOMI/08%E5%A4%A7%E6%A8%A1%E5%9E%8B%E7%AE%97%E6%B3%95/template.pptx) [chatGPT01](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E5%88%86%E5%B8%83%E5%BC%8FZOMI/08%E5%A4%A7%E6%A8%A1%E5%9E%8B%E7%AE%97%E6%B3%95/chatGPT/chatGPT01.pdf) [chatGPT02](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E5%88%86%E5%B8%83%E5%BC%8FZOMI/08%E5%A4%A7%E6%A8%A1%E5%9E%8B%E7%AE%97%E6%B3%95/chatGPT/chatGPT02.pdf) [chatGPT03](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E5%88%86%E5%B8%83%E5%BC%8FZOMI/08%E5%A4%A7%E6%A8%A1%E5%9E%8B%E7%AE%97%E6%B3%95/chatGPT/chatGPT03.pdf) - 31、[分布式ZOMI_09分布式并行系列基本介绍](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E5%88%86%E5%B8%83%E5%BC%8FZOMI/09%E5%88%86%E5%B8%83%E5%BC%8F%E5%B9%B6%E8%A1%8C%E7%B3%BB%E5%88%97%E5%9F%BA%E6%9C%AC%E4%BB%8B%E7%BB%8D.pdf) - 32、[分布式ZOMI_10分布式并行_数据并行](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E5%88%86%E5%B8%83%E5%BC%8FZOMI/10%E5%88%86%E5%B8%83%E5%BC%8F%E5%B9%B6%E8%A1%8C_%E6%95%B0%E6%8D%AE%E5%B9%B6%E8%A1%8C.pdf) - 33、[分布式ZOMI_11分布式并行_张量并行](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E5%88%86%E5%B8%83%E5%BC%8FZOMI/11%E5%88%86%E5%B8%83%E5%BC%8F%E5%B9%B6%E8%A1%8C_%E5%BC%A0%E9%87%8F%E5%B9%B6%E8%A1%8C.pdf) - 34、[分布式ZOMI_12分布式并行_流水并行](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E5%88%86%E5%B8%83%E5%BC%8FZOMI/12%E5%88%86%E5%B8%83%E5%BC%8F%E5%B9%B6%E8%A1%8C_%E6%B5%81%E6%B0%B4%E5%B9%B6%E8%A1%8C.pdf) - 35、[分布式ZOMI_13分布式并行_混合并行](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E5%88%86%E5%B8%83%E5%BC%8FZOMI/13%E5%88%86%E5%B8%83%E5%BC%8F%E5%B9%B6%E8%A1%8C_%E6%B7%B7%E5%90%88%E5%B9%B6%E8%A1%8C.pdf) - 36、[分布式ZOMI_14分布式并行_总结](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/13%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83/%E5%88%86%E5%B8%83%E5%BC%8FZOMI/14%E5%88%86%E5%B8%83%E5%BC%8F%E5%B9%B6%E8%A1%8C_%E6%80%BB%E7%BB%93.pdf) - 37、[System_for_AI Microsoft_05分布式训练算法](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/System_for_AI%20Microsoft/05%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83%E7%AE%97%E6%B3%95.pdf) - 38、[System_for_AI Microsoft_06分布式训练系统](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/System_for_AI%20Microsoft/06%E5%88%86%E5%B8%83%E5%BC%8F%E8%AE%AD%E7%BB%83%E7%B3%BB%E7%BB%9F.pdf) - 39、[System_for_AI Microsoft_07异构计算集群调度与资源管理系统](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/System_for_AI%20Microsoft/07%E5%BC%82%E6%9E%84%E8%AE%A1%E7%AE%97%E9%9B%86%E7%BE%A4%E8%B0%83%E5%BA%A6%E4%B8%8E%E8%B5%84%E6%BA%90%E7%AE%A1%E7%90%86%E7%B3%BB%E7%BB%9F.pdf) #### 14模型部署 - 1、[模型压缩_深度神经网络压缩与加速综述_纪荣嵘](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/14模型部署/综述/模型压缩_深度神经网络压缩与加速综述_纪荣嵘.pdf) - 2、[模型压缩_深度神经网络模型压缩综述_李江昀](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/14模型部署/综述/模型压缩_深度神经网络模型压缩综述_李江昀.pdf) - 3、[深度学习模型压缩与加速综述_高晗](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/14模型部署/综述/深度学习模型压缩与加速综述_高晗.pdf) - 4、[深度神经网络压缩与加速综述_曾焕强](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/14模型部署/综述/深度神经网络压缩与加速综述_曾焕强.pdf) - 5、[深度神经网络模型压缩综述_耿丽丽](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/14模型部署/综述/深度神经网络模型压缩综述_耿丽丽.pdf) - 6、[深度网络模型压缩综述_雷杰](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/14模型部署/综述/深度网络模型压缩综述_雷杰.pdf) - 7、[神经网络模型压缩方法综述_曹文龙](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/14模型部署/综述/神经网络模型压缩方法综述_曹文龙.pdf) - 8、[面向智能移动端的深度学习模型压缩方法研究_秦晴](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/14模型部署/综述/面向智能移动端的深度学习模型压缩方法研究_秦晴.pdf) - 9、[基于在线量化的深度学习模型加速技术研究_秦阳](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/14模型部署/量化/基于在线量化的深度学习模型加速技术研究_秦阳.pdf) - 10、[深度学习模型的权值交互量化算法研究_肖国麟](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/14模型部署/量化/深度学习模型的权值交互量化算法研究_肖国麟.pdf) - 11、[知识蒸馏研究综述_黄震华](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/14模型部署/知识蒸馏/知识蒸馏研究综述_黄震华.pdf) - 12、[深度学习的轻量化神经网络结构研究综述_王军](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/14模型部署/轻量化卷积核/深度学习的轻量化神经网络结构研究综述_王军.pdf) - 13、[紧凑的神经网络模型设计研究综述_郎磊](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/14模型部署/轻量化卷积核/紧凑的神经网络模型设计研究综述_郎磊.pdf) - 14、[神经网络结构搜索方法综述_刘建伟](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/14模型部署/神经结构搜索/神经网络结构搜索方法综述_刘建伟.pdf) - 15、[神经结构搜索的研究进展综述_李航宇](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/14模型部署/神经结构搜索/神经结构搜索的研究进展综述_李航宇.pdf) - 16、[Neural Architecture Search A Survey](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E7%A5%9E%E7%BB%8F%E7%BB%93%E6%9E%84%E6%90%9C%E7%B4%A2/Neural_Architecture_Search_A_Survey.pdf) - 17、[NSGA-Net Neural Architecture Search using Multi-Objective Genetic Algorithm](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E7%A5%9E%E7%BB%8F%E7%BB%93%E6%9E%84%E6%90%9C%E7%B4%A2/NSGA-Net_Neural_Architecture_Search_using_Multi-Objective_Genetic_Algorithm.pdf) - 18、[RENAS Reinforced Evolutionary Neural Architecture Search](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E7%A5%9E%E7%BB%8F%E7%BB%93%E6%9E%84%E6%90%9C%E7%B4%A2/RENAS_Reinforced_Evolutionary_Neural_Architecture_Search.pdf) - 19、[面向多核众核平台的深度学习推理加速技术研究_朱科潜](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/14模型部署/推理加速/面向多核众核平台的深度学习推理加速技术研究_朱科潜.pdf) - 20、[推理_嵌入式设备可信运行环境机器学习服务的研究与实现_刘伟浩](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/14模型部署/模型安全/推理_嵌入式设备可信运行环境机器学习服务的研究与实现_刘伟浩.pdf) - 21、[模型安全_基于联邦学习的本地模型隐私保护研究_潘凯云](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/14模型部署/模型安全/模型安全_基于联邦学习的本地模型隐私保护研究_潘凯云.pdf) - 23、[Data-free Parameter Pruning for Deep Neural Networks 数据无关的非结构化剪枝，相似性度量，减去神经元](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E9%87%8F%E5%8C%96/Data-Free_Network_Compression_via_Parametric_Non-uniform_Mixed_Precision_Quantization.pdf) - 24、[NIPS-2017-learning-to-prune-deep-neural-networks-via-layer-wise-optimal-brain-surgeon-Paper 非结构化，分层OBS](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/14模型部署/剪枝/NIPS-2017-learning-to-prune-deep-neural-networks-via-layer-wise-optimal-brain-surgeon-Paper.pdf) - 25、[NIPS-2015-learning-both-weights-and-connections-for-efficient-neural-network-Paper 非结构化，迭代的剪枝](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/14模型部署/剪枝/NIPS-2015-learning-both-weights-and-connections-for-efficient-neural-network-Paper.pdf) - 26、[NIPS-2016-dynamic-network-surgery-for-efficient-dnns-Paper 非结构化，可恢复误剪的连接](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/14模型部署/剪枝/NIPS-2016-dynamic-network-surgery-for-efficient-dnns-Paper.pdf) - 27、[NeurIPS-2018-synaptic-strength-for-convolutional-neural-network-Paper 结构化，通过突触强度决定连接重要性](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/14模型部署/剪枝/NeurIPS-2018-synaptic-strength-for-convolutional-neural-network-Paper.pdf) - 28、[SNIP SINGLE-SHOT NETWORK PRUNING BASED ON CONNECTION SENSITIVITY 非结构化，无需训练的剪枝](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/14模型部署/剪枝/SNIP_SINGLE-SHOT_NETWORK_PRUNING_BASED_ON_CONNECTION_SENSITIVITY.pdf) - 29、[Diversity Networks: Neural Network Compression Using Determinantal Point Processes 非结构化，对神经元剪枝，无需重训练](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/14模型部署/剪枝/Diversity_Networks:_Neural_Network_Compression_Using_Determinantal_Point_Processes.pdf) - 30、[NIPS-2016-perforatedcnns-acceleration-through-elimination-of-redundant-convolutions-Paper 结构化剪枝，通过掩码矩阵，可加速推理](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/14模型部署/剪枝/NIPS-2016-perforatedcnns-acceleration-through-elimination-of-redundant-convolutions-Paper.pdf) - 31、[Lebedev_Fast_ConvNets_Using_CVPR_2016_paper 结构化剪枝，组级别](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/14模型部署/剪枝/Lebedev_Fast_ConvNets_Using_CVPR_2016_paper.pdf) - 32、[PRUNING FILTERS FOR EFFICIENT CONVNETS 结构化剪枝，filter级别，l1范数决定剪枝的过滤器](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/14模型部署/剪枝/PRUNING_FILTERS_FOR_EFFICIENT_CONVNETS.pdf) - 33、[Yang_Designing_Energy-Efficient_Convolutional_CVPR_2017_paper 非结构化剪枝，一种基于能耗的剪枝方法](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/14模型部署/剪枝/Yang_Designing_Energy-Efficient_Convolutional_CVPR_2017_paper.pdf) - 34、[A Survey of Quantization Methods for Efficient Neural Network Inference 量化综述](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E9%87%8F%E5%8C%96/A_Survey_of_Quantization_Methods_for_Efficient_Neural_Network_Inference.pdf) - 35、[Dong_HAWQ_Hessian_AWare_Quantization_of_Neural_Networks_With_Mixed-Precision_ICCV_2019_paper 混合精度量化，基于Hessian矩阵，需重训练](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E9%87%8F%E5%8C%96/Dong_HAWQ_Hessian_AWare_Quantization_of_Neural_Networks_With_Mixed-Precision_ICCV_2019_paper.pdf) - 36、[MIXED-PRECISION NEURAL NETWORKS: A SURVEY 混合精度量化综述](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E9%87%8F%E5%8C%96/MIXED-PRECISION_NEURAL_NETWORKS_A_SURVEY.pdf) - 37、[SQUANT: ON-THE-FLY DATA-FREE QUANTIZATION VIA DIAGONAL HESSIAN APPROXIMATION 无数据的量化方法，对权重和激活值，效率很高 ](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E9%87%8F%E5%8C%96/squant_on_the_fly_data_free_qu.pdf) - 38、[Efficient Execution of Quantized Deep Learning Models: A Compiler Approach TVM对预量化模型的处理方法](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E9%87%8F%E5%8C%96/Efficient_Execution_of_Quantized_Deep_Learning_Models_A_Compiler_Approach.pdf) - 39、[Nagel_Data-Free_Quantization_Through_Weight_Equalization_and_Bias_Correction_ICCV_2019_paper 无数据的量化方法，调整权重的值使得网络更易于量化，大量采用BN层的信息指导优化过程](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/14模型部署/量化/Nagel_Data-Free_Quantization_Through_Weight_Equalization_and_Bias_Correction_ICCV_2019_paper.pdf) - 40、[ZeroQ: A Novel Zero Shot Quantization Framework 无数据的权重混合精度量化方法，通过BN层信息生成数据，激活值采用固定精度量化](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E9%87%8F%E5%8C%96/ZeroQ_A_Novel_Zero_Shot_Quantization_Framework.pdf) - 41、[Improving Neural Network Quantization without Retraining using Outlier Channel Splitting 量化时对权重和激活值中异常值的处理方法，将其进行分割两部分，保留异常值](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E9%87%8F%E5%8C%96/Improving_Neural_Network_Quantization_without_Retraining_using_Outlier_Channel_Splitting.pdf) - 42、[推理ZOMI_01推理系统介绍](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E6%8E%A8%E7%90%86ZOMI/01%E6%8E%A8%E7%90%86%E7%B3%BB%E7%BB%9F%E5%86%85%E5%AE%B9%E4%BB%8B%E7%BB%8D.pdf) - 43、[推理ZOMI_02推理系统与推理引擎区别](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E6%8E%A8%E7%90%86ZOMI/02%E6%8E%A8%E7%90%86%E7%B3%BB%E7%BB%9F%E4%B8%8E%E6%8E%A8%E7%90%86%E5%BC%95%E6%93%8E%E5%8C%BA%E5%88%AB.pdf) - 44、[推理ZOMI_03推理工作流](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E6%8E%A8%E7%90%86ZOMI/03%E6%8E%A8%E7%90%86%E5%B7%A5%E4%BD%9C%E6%B5%81.pdf) - 45、[推理ZOMI_04推理系统介绍](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E6%8E%A8%E7%90%86ZOMI/04%E6%8E%A8%E7%90%86%E7%B3%BB%E7%BB%9F%E4%BB%8B%E7%BB%8D.pdf) - 46、[推理ZOMI_05推理引擎特点及技术挑战](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E6%8E%A8%E7%90%86ZOMI/05%E6%8E%A8%E7%90%86%E5%BC%95%E6%93%8E%E7%89%B9%E7%82%B9%E5%8F%8A%E6%8A%80%E6%9C%AF%E6%8C%91%E6%88%98.pdf) - 47、[推理ZOMI_06推理引擎整体架构及工作流程](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E6%8E%A8%E7%90%86ZOMI/06%E6%8E%A8%E7%90%86%E5%BC%95%E6%93%8E%E6%95%B4%E4%BD%93%E6%9E%B6%E6%9E%84%E5%8F%8A%E5%B7%A5%E4%BD%9C%E6%B5%81%E7%A8%8B.pdf) - 48、[推理ZOMI_07模型小型化_推理参数介绍](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E6%8E%A8%E7%90%86ZOMI/07%E6%A8%A1%E5%9E%8B%E5%B0%8F%E5%9E%8B%E5%8C%96_%E6%8E%A8%E7%90%86%E5%8F%82%E6%95%B0%E4%BB%8B%E7%BB%8D.pdf) - 49、[推理ZOMI_08模型小型化_CNN小型化概念](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E6%8E%A8%E7%90%86ZOMI/08%E6%A8%A1%E5%9E%8B%E5%B0%8F%E5%9E%8B%E5%8C%96_CNN%E5%B0%8F%E5%9E%8B%E5%8C%96%E6%A6%82%E5%BF%B5.pdf) - 50、[推理ZOMI_09模型小型化_Transformer小型化结构](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E6%8E%A8%E7%90%86ZOMI/09%E6%A8%A1%E5%9E%8B%E5%B0%8F%E5%9E%8B%E5%8C%96_Transform%E5%B0%8F%E5%9E%8B%E5%8C%96%E7%BB%93%E6%9E%84.pdf) - 51、[推理ZOMI_10模型离线压缩_模型压缩介绍](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E6%8E%A8%E7%90%86ZOMI/10%E7%A6%BB%E7%BA%BF%E4%BC%98%E5%8C%96%E5%8E%8B%E7%BC%A9_%E6%A8%A1%E5%9E%8B%E5%8E%8B%E7%BC%A9%E4%BB%8B%E7%BB%8D.pdf) - 52、[推理ZOMI_11模型离线压缩_低比特量化基础及原理](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E6%8E%A8%E7%90%86ZOMI/11%E7%A6%BB%E7%BA%BF%E4%BC%98%E5%8C%96%E5%8E%8B%E7%BC%A9_%E4%BD%8E%E6%AF%94%E7%89%B9%E9%87%8F%E5%8C%96%E5%9F%BA%E7%A1%80%E5%8F%8A%E5%8E%9F%E7%90%86.pdf) - 53、[推理ZOMI_12模型离线压缩_低比特QAT](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E6%8E%A8%E7%90%86ZOMI/12%E7%A6%BB%E7%BA%BF%E4%BC%98%E5%8C%96%E5%8E%8B%E7%BC%A9_%E4%BD%8E%E6%AF%94%E7%89%B9%E9%87%8F%E5%8C%96QAT.pdf) - 54、[推理ZOMI_13模型离线压缩_低比特量化PTQ及部署](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E6%8E%A8%E7%90%86ZOMI/13%E7%A6%BB%E7%BA%BF%E4%BC%98%E5%8C%96%E5%8E%8B%E7%BC%A9_%E4%BD%8E%E6%AF%94%E7%89%B9%E9%87%8F%E5%8C%96PTQ%E5%8F%8A%E9%83%A8%E7%BD%B2.pdf) - 55、[推理ZOMI_14模型离线压缩_模型剪枝](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E6%8E%A8%E7%90%86ZOMI/13%E7%A6%BB%E7%BA%BF%E4%BC%98%E5%8C%96%E5%8E%8B%E7%BC%A9_%E6%A8%A1%E5%9E%8B%E5%89%AA%E6%9E%9D.pdf) - 56、[推理ZOMI_15模型离线压缩_知识蒸馏](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E6%8E%A8%E7%90%86ZOMI/14%E7%A6%BB%E7%BA%BF%E4%BC%98%E5%8C%96%E5%8E%8B%E7%BC%A9_%E7%9F%A5%E8%AF%86%E8%92%B8%E9%A6%8F.pdf) - 57、[推理ZOMI_16模型转换与优化_架构与流程](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E6%8E%A8%E7%90%86ZOMI/15%E6%A8%A1%E5%9E%8B%E8%BD%AC%E6%8D%A2%E4%B8%8E%E4%BC%98%E5%8C%96_%E6%9E%B6%E6%9E%84%E4%B8%8E%E6%B5%81%E7%A8%8B.pdf) - 58、[推理ZOMI_17模型转换与优化_模型序列化与目标生成](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E6%8E%A8%E7%90%86ZOMI/16%E6%A8%A1%E5%9E%8B%E8%BD%AC%E6%8D%A2%E4%B8%8E%E4%BC%98%E5%8C%96_%E6%A8%A1%E5%9E%8B%E5%BA%8F%E5%88%97%E5%8C%96%E4%B8%8E%E7%9B%AE%E6%A0%87%E7%94%9F%E6%88%90.pdf) - 59、[推理ZOMI_18模型转换与优化_自定义IR](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E6%8E%A8%E7%90%86ZOMI/17%E6%A8%A1%E5%9E%8B%E8%BD%AC%E6%8D%A2%E4%B8%8E%E4%BC%98%E5%8C%96_%E8%87%AA%E5%AE%9A%E4%B9%89IR.pdf) - 60、[推理ZOMI_19模型转换与优化_转换流程](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E6%8E%A8%E7%90%86ZOMI/18%E6%A8%A1%E5%9E%8B%E8%BD%AC%E6%8D%A2%E4%B8%8E%E4%BC%98%E5%8C%96_%E8%BD%AC%E6%8D%A2%E6%B5%81%E7%A8%8B.pdf) - 61、[推理ZOMI_20模型转换与优化_计算图优化](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E6%8E%A8%E7%90%86ZOMI/19%E6%A8%A1%E5%9E%8B%E8%BD%AC%E6%8D%A2%E4%B8%8E%E4%BC%98%E5%8C%96_%E8%AE%A1%E7%AE%97%E5%9B%BE%E4%BC%98%E5%8C%96.pdf) - 62、[推理ZOMI_21模型转换与优化_计算图优化详解](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E6%8E%A8%E7%90%86ZOMI/20%E6%A8%A1%E5%9E%8B%E8%BD%AC%E6%8D%A2%E4%B8%8E%E4%BC%98%E5%8C%96_%E8%AE%A1%E7%AE%97%E5%9B%BE%E4%BC%98%E5%8C%96%E8%AF%A6%E8%A7%A3.pdf) - 63、[推理ZOMI_22模型转换与优化_其他图优化](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E6%8E%A8%E7%90%86ZOMI/21%E6%A8%A1%E5%9E%8B%E8%BD%AC%E6%8D%A2%E4%B8%8E%E4%BC%98%E5%8C%96_%E5%85%B6%E4%BB%96%E5%9B%BE%E4%BC%98%E5%8C%96.pdf) - 64、[推理ZOMI_23Kernel优化_基本介绍](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E6%8E%A8%E7%90%86ZOMI/22Kernel%E4%BC%98%E5%8C%96_%E5%9F%BA%E6%9C%AC%E4%BB%8B%E7%BB%8D.pptx) - 65、[推理ZOMI_24Kernel优化_卷积优化概念](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E6%8E%A8%E7%90%86ZOMI/23Kernel%E4%BC%98%E5%8C%96_%E5%8D%B7%E7%A7%AF%E4%BC%98%E5%8C%96%E6%A6%82%E5%BF%B5.pptx) - 66、[推理ZOMI_25Kernel优化_卷积优化im2col及空间组合](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E6%8E%A8%E7%90%86ZOMI/24Kernel%E4%BC%98%E5%8C%96_%E5%8D%B7%E7%A7%AF%E4%BC%98%E5%8C%96im2col%E5%8F%8A%E7%A9%BA%E9%97%B4%E7%BB%84%E5%90%88.pptx) - 67、[推理ZOMI_26Kernel优化_卷积优化winograd](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E6%8E%A8%E7%90%86ZOMI/25Kernel%E4%BC%98%E5%8C%96_%E5%8D%B7%E7%A7%AF%E4%BC%98%E5%8C%96winograd.pptx) - 68、[推理ZOMI_27Kernel优化_间接卷积优化qnnpack](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E6%8E%A8%E7%90%86ZOMI/26Kernel%E4%BC%98%E5%8C%96_%E9%97%B4%E6%8E%A5%E5%8D%B7%E7%A7%AF%E4%BC%98%E5%8C%96qnnpack.pptx) - 69、[推理ZOMI_28Kernel优化_内存布局](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E6%8E%A8%E7%90%86ZOMI/27Kernel%E4%BC%98%E5%8C%96_%E5%86%85%E5%AD%98%E5%B8%83%E5%B1%80.pptx) - 70、[推理ZOMI_29Kernel优化_nc4hw4](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E6%8E%A8%E7%90%86ZOMI/28Kernel%E4%BC%98%E5%8C%96_nc4hw4.pptx) - 71、[推理ZOMI_30Kernel优化_汇编与循环优化及调度优化](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E6%8E%A8%E7%90%86ZOMI/29Kernel%E4%BC%98%E5%8C%96_%E6%B1%87%E7%BC%96%E4%B8%8E%E5%BE%AA%E7%8E%AF%E4%BC%98%E5%8C%96%E5%8F%8A%E8%B0%83%E5%BA%A6%E4%BC%98%E5%8C%96.pptx) - 71、[推理ZOMI_31Runtime优化_基本介绍](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/14%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/%E6%8E%A8%E7%90%86ZOMI/30Runtime%E4%BC%98%E5%8C%96%E5%9F%BA%E6%9C%AC%E4%BB%8B%E7%BB%8D.pptx) - 72、[System_for_AI Microsoft_08深度学习推导系统](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/System_for_AI%20Microsoft/08%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E6%8E%A8%E5%AF%BC%E7%B3%BB%E7%BB%9F.pdf) - 373、[System_for_AI Microsoft_10神经网络的压缩与稀疏优化](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/03%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E8%AE%BE%E8%AE%A1/System_for_AI%20Microsoft/10%E7%A5%9E%E7%BB%8F%E7%BD%91%E7%BB%9C%E7%9A%84%E5%8E%8B%E7%BC%A9%E4%B8%8E%E7%A8%80%E7%96%8F%E5%8C%96%E4%BC%98%E5%8C%96.pdf) #### 会议论文集 - 1、ATC 2022 [论文列表](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/ACT2022/ATC2022_list.pdf) [论文集](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/ACT2022/ATC2022.pdf) - 2、OSDI 2022 [论文列表](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/OSDI2022/OSDI2022_list.pdf) [论文集](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/OSDI2022/OSDI2022.pdf) - 3、ASPLOS 2022 [论文列表](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/ASPLOS2022/ASPLOS2022_list.pdf) [论文集1](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/ASPLOS2022/ASPLOS2022_1.pdf) [论文集2](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/ASPLOS2022/ASPLOS2022_2.pdf) 2023 [论文列表1](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/ASPLOS2022/ASPLOS23_1_list.pdf)[论文集1](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/ASPLOS2022/ASPLOS23_1.pdf) [论文列表2](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/ASPLOS2022/ASPLOS23_2_list.pdf)[论文集2a](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/ASPLOS2022/ASPLOS23_2a.pdf)[论文集2b](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/ASPLOS2022/ASPLOS23_2b.pdf) [论文列表3](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/ASPLOS2022/ASPLOS23_3_list.pdf)[论文集3] - 4、PLDI 2022 [论文列表](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/PLDI2022/PLDI2022_list.pdf) [论文集](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/PLDI2022/PLDI2022.pdf) - 5、POPL 2022 [论文列表](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/POPL2022/POPL2022_list.pdf) [论文集](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/POPL2022/POPL2022.pdf) - 6、CGO 2022 [论文列表](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/CGO2022/CGO2022_list.pdf) [论文集](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/CGO2022/CGO2022.pdf) - 7、MLSys 2022 [论文列表/集](https://gitee.com/wanglei07/DeepLearningCompiling/blob/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/MLSys2022/MLSys2022_list.md) - 8、PACT 2022 [论文列表](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/PACT2022/pact22_list.pdf)[论文集](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/PACT2022/pact22.pdf) - 9、SC 2022 [论文列表](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/SC2022/SC2022_list.pdf) [论文集1](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/SC2022/SC2022_1.pdf) [论文集2](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/SC2022/SC2022_2.pdf) - 10、PPoPP 2022 [论文列表](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/PPoPP2022/PPoPP_list.pdf) [论文集](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/PPoPP2022/PPOPP22.pdf) - 11、ISCA 2022 [论文列表](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/ISCA2022/ISCA_list.pdf) [论文集1](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/ISCA2022/ISCA22_1.pdf) [论文集2](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/ISCA2022/ISCA22_2.pdf) - 12、OOPSLA 2022 [论文列表1](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/OOPSLA2022/OOPLSA1_list.pdf) [论文列表2](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/OOPSLA2022/OOPLSA2_list.pdf) [论文集1](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/OOPSLA2022/OOPLSA1.pdf) [论文集2](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/OOPSLA2022/OOPLSA2.pdf) - 13、TOPLAS 2022 [论文集1](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/TOPLAS2022/TOPLAS_1.pdf) [论文集2](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/TOPLAS2022/TOPLAS_2.pdf) [论文集3](https://gitee.com/wanglei07/DeepLearningCompiling/raw/master/%E4%BC%9A%E8%AE%AE%E8%AE%BA%E6%96%87%E9%9B%86/TOPLAS2022/TOPLAS_3.pdf)