高性能，深度学习

跨多个平台加速深度神经网络工作负载，以实现更快、更准确的人工智能推理结果。

如何工作

逐步完成使用英特尔® Distribution of OpenVINO™ 工具套件的流程，并仔细研究从设置和计划解决方案到部署的关键步骤。

模型优化工具

优化模型

模型优化器是基于 Python* 的工具，用于导入训练有素的模型并将其转换为中间表示形式。该工具通过保守的拓扑转换来优化性能或空间。完成的优化与硬件无关。

模型优化器简介

可视化和分析

充分利用深度学习工作台，它是一个基于 Web 的图形环境，使用户可以可视化、微调和比较深度学习模型在各种目标环境上的性能。

深度学习工作台

性能微调

训练后优化工具通过将模型转换为更加硬件友好的表示形式来加快推理速度。它通过应用无需重新训练的方法来做到这一点。

优化最佳实践

训练后优化工具

准备开始了吗？

挑选并下载

产品和性能信息

¹

性能因用途、配置和其他因素而异。请访问 www.Intel.com/PerformanceIndex 了解更多信息。

提供反馈

<script>var isTargetEnabled =false;</script>

<script type="text/javascript">
    var cq_tms = {
        wa_emtcontenttype: "emtcontenttype:webpage/landingpage",
        wa_rsoftware: "rsoftware:developmenttools,rsoftware:componentsproducts/inteldistributionofopenvinotoolkit",
        wa_emtsubject: "emtsubject:itinformationtechnology/aiartificialintelligence/deeplearning",
        
        wa_industry_type: "",
        wa_partner_prog: "",
        wa_subject: "emtsubject:itinformationtechnology/aiartificialintelligence/deeplearning",
        wa_biz_size: "",
        wa_emt_org: "",
        wa_emt_intel: "",
        wa_partner_type: "",
        wa_programidentifier: "idz",
        wa_location: "cn",
        wa_language: "zh",
        wa_english_title: "Deep Learning Optimization",
        wa_content_type: "landingpage",
        wa_intel_platform: "",
        wa_applications: "",
        wa_local: "zh_CN",
        wa_intel_technology: "",
        wa_system_type: "",
        wa_product_name: "",
        wa_audience: "emtaudience:business/btssbusinesstechnologysolutionspecialist/softwaredeveloper",
        wa_life_cycle: "",
        wa_cq_url: "/content/www/cn/zh/develop/tools/openvino-toolkit/optimization.html",
        wa_page_type_micro: "mf-contentdetailpage",
        wa_env: "prd",
        wa_cq_pub_env: "publish",
        wa_transl_status: "1",
        wa_product_id: "",
        wa_software: "rsoftware:componentsproducts/inteldistributionofopenvinotoolkit,rsoftware:developmenttools",
        wa_reference_design: "",
        wa_codename: "",
        wa_created_date: "1602026678676",
        wa_modifed_date: "1610764066626",
        wa_off_time: "1633188780000",
        wa_ownedby: "",
        wa_managedby: "",
        wa_idz_nid : "",
        wa_event_type: "",
        wa_idz_custom_tags: "",
        wa_programming_language: "",
        wa_skill_level: "",
        wa_solution: "",
        wa_product_formFactor: "",
        wa_operating_system: "",
        wa_marketing_products: "",
        wa_ssg_third_party_prod_sys_sol: "",
        
        wa_secondary_content_type: ""
    };
    
    
    
    cq_tms.wa_tms_sync_load = "deferred";
    
</script>

<script type="text/javascript" src="/content/dam/www/global/wap/tms.sync.js" async></script>

<link rel="stylesheet" href="/etc.clientlibs/settings/wcm/designs/ver/201210/intel/clientlibs/pages/commons-page.min.css" type="text/css"><script src="/etc.clientlibs/settings/wcm/designs/ver/201210/intel/clientlibs/pages/commons-page.min.js" defer></script>

<link rel="stylesheet" href="/etc.clientlibs/settings/wcm/designs/ver/201210/intel/clientlibs/pages/contact-us.min.css" type="text/css">

<script>!function(){var e=setInterval(function(){"undefined"!=typeof $CQ&&($CQ(function(){CQ_Analytics.SegmentMgr.loadSegments("/etc/segmentation"),CQ_Analytics.ClientContextUtils.init("/etc/clientcontext/intel",window.location.pathname.substr(0,window.location.pathname.indexOf(".")))}),clearInterval(e))},100)}();</script>

<link rel="stylesheet" type="text/css" href="/etc.clientlibs/settings/wcm/designs/intel/cn/zh/css/resources/css/intel.rwd.override.css"/>