[{"data":1,"prerenderedAt":450},["ShallowReactive",2],{"content-query-qDEm0BJL2I":3},{"_path":4,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":8,"description":9,"date":10,"cover":11,"type":12,"body":13,"_type":444,"_id":445,"_source":446,"_file":447,"_stem":448,"_extension":449},"/news/zh/1633","zh",false,"","7月7日易用性SIG技术分享活动精彩回顾","加入易用性SIG交流群，共同打造易学易用、灵活高效的AI框架！","2022-07-12","https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2022/07/18/484daa36697b4adabe5ee0244d2f4928.png","news",{"type":14,"children":15,"toc":441},"root",[16,24,33,53,61,73,85,92,100,119,131,141,151,158,168,178,183,190,200,210,215,222,230,240,245,250,255,264,269,274,279,288,293,298,303,311,321,332,342,346,354,359,366,374,384,400,410,426],{"type":17,"tag":18,"props":19,"children":21},"element","h1",{"id":20},"_7月7日易用性sig技术分享活动精彩回顾",[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/2022/07/18/99174260501a403895a320bfd9a31cb4.gif",[],{"type":17,"tag":25,"props":34,"children":35},{},[36],{"type":17,"tag":37,"props":38,"children":39},"strong",{},[40,46,48],{"type":17,"tag":41,"props":42,"children":43},"em",{},[44],{"type":23,"value":45},"易用性SIG",{"type":23,"value":47},"****",{"type":17,"tag":41,"props":49,"children":50},{},[51],{"type":23,"value":52},"技术分享",{"type":17,"tag":25,"props":54,"children":55},{},[56],{"type":17,"tag":37,"props":57,"children":58},{},[59],{"type":23,"value":60},"精彩回顾",{"type":17,"tag":25,"props":62,"children":63},{},[64,66,71],{"type":23,"value":65},"2022年7月7日，MindSpore易用性SIG举行了第3次线上技术分享活动，主题是**《MindSpore在CV领域的应用案例》",{"type":17,"tag":37,"props":67,"children":68},{},[69],{"type":23,"value":70},"，共有",{"type":23,"value":72},"150+**位开发者朋友通过腾讯会议和B站直播参与了本次活动。",{"type":17,"tag":25,"props":74,"children":75},{},[76,78,83],{"type":23,"value":77},"此次活动新吸纳了",{"type":17,"tag":37,"props":79,"children":80},{},[81],{"type":23,"value":82},"45",{"type":23,"value":84},"位开发者加入易用性SIG交流群。感谢开发者们的热情参与和各位老师的精彩分享，在此我们整理了此次活动的内容摘要以及提问和回答，希望大家满意！",{"type":17,"tag":25,"props":86,"children":87},{},[88],{"type":17,"tag":29,"props":89,"children":91},{"alt":7,"src":90},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2022/07/18/c5acf6bf554d4d33ade392a34cbd71ff.png",[],{"type":17,"tag":25,"props":93,"children":94},{},[95],{"type":17,"tag":37,"props":96,"children":97},{},[98],{"type":23,"value":99},"分享内容回顾",{"type":17,"tag":25,"props":101,"children":102},{},[103,105,110,112,117],{"type":23,"value":104},"本次活动的主题是**《MindSpore在CV领域的应用案例》",{"type":17,"tag":37,"props":106,"children":107},{},[108],{"type":23,"value":109},"，3位演讲嘉宾面向MindSpore初学者和有一定经验的使用者，分享",{"type":23,"value":111},"MindSpore的自动****、手动安装",{"type":17,"tag":37,"props":113,"children":114},{},[115],{"type":23,"value":116},"和",{"type":23,"value":118},"MindSpore在CV领域的应用案例**。",{"type":17,"tag":25,"props":120,"children":121},{},[122,124,129],{"type":23,"value":123},"感谢",{"type":17,"tag":37,"props":125,"children":126},{},[127],{"type":23,"value":128},"彭博、沈阳、甜粽",{"type":23,"value":130},"为大家带来的精彩分享！",{"type":17,"tag":25,"props":132,"children":133},{},[134,139],{"type":17,"tag":37,"props":135,"children":136},{},[137],{"type":23,"value":138},"01",{"type":23,"value":140},"**《基于MindSpore的深度学习模型在多媒体领域的实践》**",{"type":17,"tag":25,"props":142,"children":143},{},[144,149],{"type":17,"tag":37,"props":145,"children":146},{},[147],{"type":23,"value":148},"复旦大学计算机科学与技术学院博士生彭博",{"type":23,"value":150},"，围绕自己近期的工作，详细介绍了基于MindSpore深度图像修复技术实践和基于MindSpore的多模态技术在纺织工业的应用，同时，为大家分享了自己的建模思路。",{"type":17,"tag":25,"props":152,"children":153},{},[154],{"type":17,"tag":29,"props":155,"children":157},{"alt":7,"src":156},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2022/07/18/aa05581a47b84a208d479168b31f28b8.png",[],{"type":17,"tag":25,"props":159,"children":160},{},[161,166],{"type":17,"tag":37,"props":162,"children":163},{},[164],{"type":23,"value":165},"02",{"type":23,"value":167},"**《基于MindSpore的细粒度哈希学习》**",{"type":17,"tag":25,"props":169,"children":170},{},[171,176],{"type":17,"tag":37,"props":172,"children":173},{},[174],{"type":23,"value":175},"南京理工大学沈阳博士生沈阳",{"type":23,"value":177},"，基于自身的研究工作，为大家介绍了什么是细粒度哈希、为什么需要细粒度哈希，以及哈希深度学习的关键技术等。",{"type":17,"tag":25,"props":179,"children":180},{},[181],{"type":23,"value":182},"同时，他详细介绍了如何基于MindSpore实现大规模细粒度图像检索的哈希学习应用案例。",{"type":17,"tag":25,"props":184,"children":185},{},[186],{"type":17,"tag":29,"props":187,"children":189},{"alt":7,"src":188},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2022/07/18/877f19c123e8496ab2a5f9171ed87033.png",[],{"type":17,"tag":25,"props":191,"children":192},{},[193,198],{"type":17,"tag":37,"props":194,"children":195},{},[196],{"type":23,"value":197},"03",{"type":23,"value":199},"**《MindSpore1.7在Windows和Linux环境下的安装》**",{"type":17,"tag":25,"props":201,"children":202},{},[203,208],{"type":17,"tag":37,"props":204,"children":205},{},[206],{"type":23,"value":207},"MindSpore爱好者甜粽同学",{"type":23,"value":209},"，在线为大家展示了在Linux环境下如何实现MindSpore的自动安装。",{"type":17,"tag":25,"props":211,"children":212},{},[213],{"type":23,"value":214},"同时，她还介绍了在Windows环境下如何实现MindSpore的pip、conda和source的手动安装。",{"type":17,"tag":25,"props":216,"children":217},{},[218],{"type":17,"tag":29,"props":219,"children":221},{"alt":7,"src":220},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2022/07/18/2c536cea62324b83b7ff0379d87e42e9.png",[],{"type":17,"tag":25,"props":223,"children":224},{},[225],{"type":17,"tag":37,"props":226,"children":227},{},[228],{"type":23,"value":229},"Q&A环节",{"type":17,"tag":25,"props":231,"children":232},{},[233,235],{"type":23,"value":234},"Q：",{"type":17,"tag":37,"props":236,"children":237},{},[238],{"type":23,"value":239},"请问彭老师，图像修复过程中的耗时最长的是哪里？有什么好的性能优化办法？",{"type":17,"tag":25,"props":241,"children":242},{},[243],{"type":23,"value":244},"A：我们设计这个模型实际上是一个端到端的模型，我还没有就哪一块去单独看它的耗时。",{"type":17,"tag":25,"props":246,"children":247},{},[248],{"type":23,"value":249},"我理解应该是，在编解码器过程中耗时最多的，因为他的参数量主要在这里，其他的地方像着色这块，只是对它的每一层的特征去做一个约束，这块的计算量应该是比较小的，主要是在生成器这块。",{"type":17,"tag":25,"props":251,"children":252},{},[253],{"type":23,"value":254},"如果我们想要优化的话，建议尽量压缩一下网络的参数量。",{"type":17,"tag":25,"props":256,"children":257},{},[258,259],{"type":23,"value":234},{"type":17,"tag":37,"props":260,"children":261},{},[262],{"type":23,"value":263},"请问沈博，使用MindSpore做科研项目过程中有什么心得或经验？",{"type":17,"tag":25,"props":265,"children":266},{},[267],{"type":23,"value":268},"A：我觉得做科研的过程，最主要的是找到一个比较细的研究问题，有了这个研究问题之后，我们研究之后会有一个想法，得到这个想法之后，通过不同的框架去建模，在建模过程中，我们可能会觉得无论是用MindSpore还是TensorFlow、 PyTorch，本质上是同一个过程。",{"type":17,"tag":25,"props":270,"children":271},{},[272],{"type":23,"value":273},"因为我们通过这些框架的使用，可能会在参数调优上有些不同，但是只要调参得当，方法正确，都会得到一个接近SOTA或超越SOTA的一个性能，更关键的就在于怎样在最初找到一个比较好的研究方向，或者说一个比较容易出成果的科研方向。",{"type":17,"tag":25,"props":275,"children":276},{},[277],{"type":23,"value":278},"比如说我们现在做的一些普通的论文，它的数据集是比较小的，可能框架的影响并没有那么大。",{"type":17,"tag":25,"props":280,"children":281},{},[282,283],{"type":23,"value":234},{"type":17,"tag":37,"props":284,"children":285},{},[286],{"type":23,"value":287},"作为小白，在安装方面，有啥好的经验和建议吗？",{"type":17,"tag":25,"props":289,"children":290},{},[291],{"type":23,"value":292},"A：如果说电脑上之前是比较空白的一个裸机，就像Python 、conda都没有接触过，都没有装的话，我比较推荐自动安装。",{"type":17,"tag":25,"props":294,"children":295},{},[296],{"type":23,"value":297},"如果已经安装了部分的依赖项，像Python3.7.5、conda，像CMake、git还有wheel这些 手动安装是比较合适的。",{"type":17,"tag":25,"props":299,"children":300},{},[301],{"type":23,"value":302},"总体来讲，自动安装、手动安装只要你肯花时间，难度都没那么大，大家一起去摸索，探索一下，可能遇到不一样的风景。",{"type":17,"tag":25,"props":304,"children":305},{},[306],{"type":17,"tag":37,"props":307,"children":308},{},[309],{"type":23,"value":310},"资料获取",{"type":17,"tag":25,"props":312,"children":313},{},[314,319],{"type":17,"tag":37,"props":315,"children":316},{},[317],{"type":23,"value":318},"1",{"type":23,"value":320}," 本次活动的演讲材料将发布到MindSpore社区，请点击链接获取（点击“阅读原文”直接进入）。",{"type":17,"tag":25,"props":322,"children":323},{},[324],{"type":17,"tag":325,"props":326,"children":330},"a",{"href":327,"rel":328},"https://gitee.com/mindspore/community/tree/master/sigs/usability",[329],"nofollow",[331],{"type":23,"value":327},{"type":17,"tag":25,"props":333,"children":334},{},[335,340],{"type":17,"tag":37,"props":336,"children":337},{},[338],{"type":23,"value":339},"2",{"type":23,"value":341}," 本次活动的视频将上传到B站MindSpore官方账号，敬请关注。",{"type":17,"tag":25,"props":343,"children":344},{},[345],{"type":23,"value":9},{"type":17,"tag":25,"props":347,"children":348},{},[349],{"type":17,"tag":37,"props":350,"children":351},{},[352],{"type":23,"value":353},"加入方式：",{"type":17,"tag":25,"props":355,"children":356},{},[357],{"type":23,"value":358},"扫码添加小助手的微信（vx: mindspore0328），添加时请备注：易用性。小助手拉你进群哦！",{"type":17,"tag":25,"props":360,"children":361},{},[362],{"type":17,"tag":29,"props":363,"children":365},{"alt":7,"src":364},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2022/07/18/5e4453ec0c7148c19b134a310bae22be.jpg",[],{"type":17,"tag":25,"props":367,"children":368},{},[369],{"type":17,"tag":37,"props":370,"children":371},{},[372],{"type":23,"value":373},"MindSpore官方资料",{"type":17,"tag":25,"props":375,"children":376},{},[377,382],{"type":17,"tag":37,"props":378,"children":379},{},[380],{"type":23,"value":381},"官方QQ群",{"type":23,"value":383}," : 486831414",{"type":17,"tag":25,"props":385,"children":386},{},[387,392,394],{"type":17,"tag":37,"props":388,"children":389},{},[390],{"type":23,"value":391},"官网",{"type":23,"value":393},"：",{"type":17,"tag":325,"props":395,"children":398},{"href":396,"rel":397},"https://www.mindspore.cn/",[329],[399],{"type":23,"value":396},{"type":17,"tag":25,"props":401,"children":402},{},[403,408],{"type":17,"tag":37,"props":404,"children":405},{},[406],{"type":23,"value":407},"Gitee",{"type":23,"value":409}," : https : //gitee.com/mindspore/mindspore",{"type":17,"tag":25,"props":411,"children":412},{},[413,418,420],{"type":17,"tag":37,"props":414,"children":415},{},[416],{"type":23,"value":417},"GitHub",{"type":23,"value":419}," : ",{"type":17,"tag":325,"props":421,"children":424},{"href":422,"rel":423},"https://github.com/mindspore-ai/mindspore",[329],[425],{"type":23,"value":422},{"type":17,"tag":25,"props":427,"children":428},{},[429,434,435],{"type":17,"tag":37,"props":430,"children":431},{},[432],{"type":23,"value":433},"论坛",{"type":23,"value":393},{"type":17,"tag":325,"props":436,"children":439},{"href":437,"rel":438},"https://bbs.huaweicloud.com/forum/forum-1076-1.html",[329],[440],{"type":23,"value":437},{"title":7,"searchDepth":442,"depth":442,"links":443},4,[],"markdown","content:news:zh:1633.md","content","news/zh/1633.md","news/zh/1633","md",1776506052376]