[{"data":1,"prerenderedAt":404},["ShallowReactive",2],{"content-query-xFkiW0Bovu":3},{"_path":4,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":8,"description":9,"date":10,"cover":11,"type":12,"body":13,"_type":398,"_id":399,"_source":400,"_file":401,"_stem":402,"_extension":403},"/news/zh/1512","zh",false,"","MindSpore初学教程 | 简单易懂的新手入门指导，深度学习拦路虎都退退退！","文档教程不直观？示例版本不够新？安装依赖捋不清？讲解视频找不到？这些问题都退退退！","2022-05-27","https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2022/06/13/c640f20846e141e089b6fa5600ee456a.png","news",{"type":14,"children":15,"toc":395},"root",[16,24,33,42,47,57,64,69,76,88,96,101,108,113,122,130,135,150,155,160,168,173,178,186,191,196,204,209,214,228,233,241,246,251,265,270,275,280,285,292,297,302,313,320,328,338,354,364,380],{"type":17,"tag":18,"props":19,"children":21},"element","h1",{"id":20},"mindspore初学教程-简单易懂的新手入门指导深度学习拦路虎都退退退",[22],{"type":23,"value":8},"text",{"type":17,"tag":25,"props":26,"children":27},"p",{},[28],{"type":17,"tag":29,"props":30,"children":32},"img",{"alt":7,"src":31},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2022/05/30/f2f3070e5dbc431dba3e844a048c25b7.gif",[],{"type":17,"tag":25,"props":34,"children":35},{},[36],{"type":17,"tag":37,"props":38,"children":39},"strong",{},[40],{"type":23,"value":41},"昇思MindSpore",{"type":17,"tag":25,"props":43,"children":44},{},[45],{"type":23,"value":46},"昇思MindSpore是一个全场景深度学习框架，已经成为人工智能领域非常流行的开源AI框架。你可能已经听说过它很多次，但一直没有去接触了解过。跟随时代发展，无论你是大牛还是小白，学习MindSpore都是很有意义的。",{"type":17,"tag":25,"props":48,"children":49},{},[50,52],{"type":23,"value":51},"作为新手，那么问题来了，文档教程不直观？示例版本不够新？安装依赖捋不清？讲解视频找不到？这些问题都",{"type":17,"tag":37,"props":53,"children":54},{},[55],{"type":23,"value":56},"退退退****！",{"type":17,"tag":25,"props":58,"children":59},{},[60],{"type":17,"tag":29,"props":61,"children":63},{"alt":7,"src":62},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2022/05/30/a306f522d40d47d79b3bb108be19b237.png",[],{"type":17,"tag":25,"props":65,"children":66},{},[67],{"type":23,"value":68},"**《MindSpore初学教程》**就是为你准备的！九节课，全视频，手把手，零起点，案例讲解，基于最新的版本。跟着下面的引导快速了解章节内容，快来一起学习进步吧~",{"type":17,"tag":25,"props":70,"children":71},{},[72],{"type":17,"tag":29,"props":73,"children":75},{"alt":7,"src":74},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2022/05/30/0873c1ca5b8c4a6aa9661948f8db26f1.png",[],{"type":17,"tag":25,"props":77,"children":78},{},[79],{"type":17,"tag":80,"props":81,"children":85},"a",{"href":82,"rel":83},"https://mp.weixin.qq.com/s/ZooW3do5f2NCGBIYc9m3sA",[84],"nofollow",[86],{"type":23,"value":87},"请在微信内打开进行观看",{"type":17,"tag":25,"props":89,"children":90},{},[91],{"type":17,"tag":37,"props":92,"children":93},{},[94],{"type":23,"value":95},"第一节 基本介绍",{"type":17,"tag":25,"props":97,"children":98},{},[99],{"type":23,"value":100},"正式进入框架的学习，首先通过整体介绍昇思MindSpore和华为昇腾AI全栈，并介绍昇思MindSpore在华为昇腾AI全栈中的位置，包括总体架构、设计理念和层次结构。从基础概念开始，对昇思MindSpore有一个系统性的认知和了解，为我们接下来的学习做好铺垫打好基础。",{"type":17,"tag":25,"props":102,"children":103},{},[104],{"type":17,"tag":29,"props":105,"children":107},{"alt":7,"src":106},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2022/05/30/fadfb30df3274d94ac4dcb60ea1d84b5.png",[],{"type":17,"tag":25,"props":109,"children":110},{},[111],{"type":23,"value":112},"本节教程的最后还会带大家体验下载安装流程，解读安装教程的基本逻辑和概念，帮助初学迷茫的新手选择合适版本。当然啦，各种详细的安装方法大家可以在官网快速查看！",{"type":17,"tag":25,"props":114,"children":115},{},[116],{"type":17,"tag":80,"props":117,"children":120},{"href":118,"rel":119},"https://www.mindspore.cn/install",[84],[121],{"type":23,"value":118},{"type":17,"tag":25,"props":123,"children":124},{},[125],{"type":17,"tag":37,"props":126,"children":127},{},[128],{"type":23,"value":129},"第二节 快速入门",{"type":17,"tag":25,"props":131,"children":132},{},[133],{"type":23,"value":134},"学习本节贯穿昇思MindSpore深度学习的基本流程，首先学习的知识点有数据的准备，创建网络模型，定义损失器与优化器，训练与推理。然后以LeNet5网络模型为例子，实现深度学习中的常见任务，带领大家一步步的去了解昇思MindSpore。",{"type":17,"tag":25,"props":136,"children":137},{},[138,143,145],{"type":17,"tag":37,"props":139,"children":140},{},[141],{"type":23,"value":142},"第三节",{"type":23,"value":144}," ",{"type":17,"tag":37,"props":146,"children":147},{},[148],{"type":23,"value":149},"张量Tensor",{"type":17,"tag":25,"props":151,"children":152},{},[153],{"type":23,"value":154},"张量是一种特殊的数据结构，与数组和矩阵非常相似，它也是昇思MindSpore网络运算中的基本数据结构。",{"type":17,"tag":25,"props":156,"children":157},{},[158],{"type":23,"value":159},"本节主要介绍张量和稀疏张量的属性及用法——首先通过了解初始化，属性，索引，转换，学习张量的基本概念；然后ZOMI酱会详细讲解MindSpore Tensor的作用，最后介绍如何构建、索引和张量进行运算。",{"type":17,"tag":25,"props":161,"children":162},{},[163],{"type":17,"tag":37,"props":164,"children":165},{},[166],{"type":23,"value":167},"第四节 数据处理",{"type":17,"tag":25,"props":169,"children":170},{},[171],{"type":23,"value":172},"mindspore.dataset提供了部分常用数据集和标准格式数据集的加载接口，使得用户能够快速进行数据处理操作。对于图像数据集，用户可使用mindvision.dataset来加载和处理数据集。",{"type":17,"tag":25,"props":174,"children":175},{},[176],{"type":23,"value":177},"本节首先介绍如何使用mindvision.dataset.Cifar10接口加载并处理CIFAR-10数据集，然后介绍如何使用mindspore.dataset.GeneratorDataset实现自定义数据集加载。",{"type":17,"tag":25,"props":179,"children":180},{},[181],{"type":17,"tag":37,"props":182,"children":183},{},[184],{"type":23,"value":185},"第五节 创建网络",{"type":17,"tag":25,"props":187,"children":188},{},[189],{"type":23,"value":190},"神经网络模型由多个数据操作层组成，mindspore.nn提供了各种网络基础模块。",{"type":17,"tag":25,"props":192,"children":193},{},[194],{"type":23,"value":195},"本节以构建LeNet-5网络为例，先展示使用mindspore.nn建立神经网络模型，再展示使用mindvision.classification.models快速构建LeNet-5网络模型。",{"type":17,"tag":25,"props":197,"children":198},{},[199],{"type":17,"tag":37,"props":200,"children":201},{},[202],{"type":23,"value":203},"第六节 自动微分",{"type":17,"tag":25,"props":205,"children":206},{},[207],{"type":23,"value":208},"自动微分能够计算可导函数在某点处的导数值，是反向传播算法的一般化。自动微分主要解决的问题是将一个复杂的数学运算分解为一系列简单的基本运算，该功能对用户屏蔽了大量的求导细节和过程，大大降低了框架的使用门槛。",{"type":17,"tag":25,"props":210,"children":211},{},[212],{"type":23,"value":213},"本节ZOMI酱会通过对数据进行求导，梯度缩放和停止计算微分这三个内容进行讲解，展开介绍一下昇思MindSpore是如何进行自动微分的。",{"type":17,"tag":25,"props":215,"children":216},{},[217,222,223],{"type":17,"tag":37,"props":218,"children":219},{},[220],{"type":23,"value":221},"第七节",{"type":23,"value":144},{"type":17,"tag":37,"props":224,"children":225},{},[226],{"type":23,"value":227},"模型训练",{"type":17,"tag":25,"props":229,"children":230},{},[231],{"type":23,"value":232},"通过前面章节的学习，相信大家已经学会如何创建模型和构建数据集。本节就是学习如何设置超参和优化模型参数，学习损失函数和优化器的定义和使用，最后基于昇思MindSpore训练一个网络模型。",{"type":17,"tag":25,"props":234,"children":235},{},[236],{"type":17,"tag":37,"props":237,"children":238},{},[239],{"type":23,"value":240},"第八节 保存加载",{"type":17,"tag":25,"props":242,"children":243},{},[244],{"type":23,"value":245},"上一章节主要介绍了如何调整超参数，并进行网络模型训练。训练网络模型的过程中，实际上我们希望保存中间和最后的结果，用于微调（fine-tune）和后续的模型部署和推理。",{"type":17,"tag":25,"props":247,"children":248},{},[249],{"type":23,"value":250},"本章节ZOMI酱就会给大家讲解在昇思MindSpore中如何对网络模型权重进行保存，训练完模型后如何加载模型权重，最后如何把网络模型用于迁移学习或者下游任务。",{"type":17,"tag":25,"props":252,"children":253},{},[254,259,260],{"type":17,"tag":37,"props":255,"children":256},{},[257],{"type":23,"value":258},"第九节",{"type":23,"value":144},{"type":17,"tag":37,"props":261,"children":262},{},[263],{"type":23,"value":264},"推理部署",{"type":17,"tag":25,"props":266,"children":267},{},[268],{"type":23,"value":269},"初学教程的最后一节，是时候放（并）大（没）招（有）啦~",{"type":17,"tag":25,"props":271,"children":272},{},[273],{"type":23,"value":274},"本章节会使用MindSpore Vision中的mobilenet_v2网络微调的方法来开发一个AI应用（狗和牛角包分类），并将训练好的网络模型部署到Android手机上，执行推理和部署功能。",{"type":17,"tag":25,"props":276,"children":277},{},[278],{"type":23,"value":279},"端到端的串联之前章节的所学内容，从网络模型的导出，到转换成端侧可部署的格式，最后真正把昇思MindSpore端云协同打通去部署和体验。",{"type":17,"tag":25,"props":281,"children":282},{},[283],{"type":23,"value":284},"贯穿昇思MindSpore的基础功能，代码解析、环境构建，执行推理…完整的开发流程体验，ZOMI酱老师每一步都做了通俗易懂的讲解，学习起来没压力！手把手的教程，大家可以动手实操一下，感受一下昇思MindSpore 的魅力。",{"type":17,"tag":25,"props":286,"children":287},{},[288],{"type":17,"tag":29,"props":289,"children":291},{"alt":7,"src":290},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2022/05/30/8d3aaccc17ac49e38a27fd5d3fec1104.png",[],{"type":17,"tag":25,"props":293,"children":294},{},[295],{"type":23,"value":296},"从基本原理到模型开发实践，讲解昇思MindSpore开源AI计算框架的基本知识和编程方法，以及使用昇思MindSpore进行实践，助力新手小白快速上手昇思MindSpore，进行AI应用开发。",{"type":17,"tag":25,"props":298,"children":299},{},[300],{"type":23,"value":301},"对昇思MindSpore感兴趣的开发者，欢迎加入昇思MindSpore的社区并一键三连（Watch/Star/Fork）~有任何问题，可以通过社区issue进行提问，直接和昇思MindSpore工程师对话交流，手把手指导解决难题！",{"type":17,"tag":25,"props":303,"children":304},{},[305,307],{"type":23,"value":306},"社区issue：",{"type":17,"tag":80,"props":308,"children":311},{"href":309,"rel":310},"https://gitee.com/mindspore/mindspore/issues",[84],[312],{"type":23,"value":309},{"type":17,"tag":25,"props":314,"children":315},{},[316],{"type":17,"tag":29,"props":317,"children":319},{"alt":7,"src":318},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2022/05/30/c5495d656e7b4cbf8d665d7f9ad95fd8.jpg",[],{"type":17,"tag":25,"props":321,"children":322},{},[323],{"type":17,"tag":37,"props":324,"children":325},{},[326],{"type":23,"value":327},"MindSpore官方资料",{"type":17,"tag":25,"props":329,"children":330},{},[331,336],{"type":17,"tag":37,"props":332,"children":333},{},[334],{"type":23,"value":335},"官方QQ群",{"type":23,"value":337}," : 486831414",{"type":17,"tag":25,"props":339,"children":340},{},[341,346,348],{"type":17,"tag":37,"props":342,"children":343},{},[344],{"type":23,"value":345},"官网",{"type":23,"value":347},"：",{"type":17,"tag":80,"props":349,"children":352},{"href":350,"rel":351},"https://www.mindspore.cn/",[84],[353],{"type":23,"value":350},{"type":17,"tag":25,"props":355,"children":356},{},[357,362],{"type":17,"tag":37,"props":358,"children":359},{},[360],{"type":23,"value":361},"Gitee",{"type":23,"value":363}," : https : //gitee.com/mindspore/mindspore",{"type":17,"tag":25,"props":365,"children":366},{},[367,372,374],{"type":17,"tag":37,"props":368,"children":369},{},[370],{"type":23,"value":371},"GitHub",{"type":23,"value":373}," : ",{"type":17,"tag":80,"props":375,"children":378},{"href":376,"rel":377},"https://github.com/mindspore-ai/mindspore",[84],[379],{"type":23,"value":376},{"type":17,"tag":25,"props":381,"children":382},{},[383,388,389],{"type":17,"tag":37,"props":384,"children":385},{},[386],{"type":23,"value":387},"论坛",{"type":23,"value":347},{"type":17,"tag":80,"props":390,"children":393},{"href":391,"rel":392},"https://bbs.huaweicloud.com/forum/forum-1076-1.html",[84],[394],{"type":23,"value":391},{"title":7,"searchDepth":396,"depth":396,"links":397},4,[],"markdown","content:news:zh:1512.md","content","news/zh/1512.md","news/zh/1512","md",1776506049546]