Technical Article

Оптимизировали, оптимизировали, да не выоптимизировали!

Оптимизация? Конечно, каждый сталкивался с данной задачей при разработке своих, сколь-нибудь значительных, требующих определённых вычислений, приложений. При этом способов оптимизировать код существует огромное множество, и, как следствие, различных путей сделать это в автоматическом режиме с помощью опций компилятора. Вот здесь и возникает проблема – как выбрать то, что нужно нам и не запутаться?

Traffic Light Detection Using the TensorFlow* Object Detection API

Abstract

This case study evaluates the ability of the TensorFlow* Object Detection API to solve a real-time problem such as traffic light detection. The experiment uses the Microsoft Common Objects in Context (COCO) pre-trained model called Single Shot Multibox Detector MobileNet from the TensorFlow Zoo for transfer learning. Intel® Xeon® and Intel® Xeon PhiTM processor-based machines were used for the study. At the end of this experiment, we obtained an accurate model that was able to identify the traffic signals at more than 90 percent accuracy.

  • Professional
  • Professors
  • Students
  • Linux*
  • Artificial Intelligence
  • Intermediate
  • Artificial Intelligence
  • Machine Learning
  • Dynamic Device Personalization for Intel® Ethernet 700 Series

    The Intel Ethernet 700 Series network adapters enable Dynamic Device Personalization (DDP). This article shows how to to program and configure DDP profiles using the DPDK, with the GTPv1 profile as an example.
  • Networking
  • Network Functions Virtualization (NFV)
  • Software Defined Networking (SDN)
  • Using Intel® Optane™ Technology with Ceph* to Build High-Performance OLTP Solutions

    This white paper describes an All-Flash Array reference architecture for OLTP workloads using Intel® Optane™ technology and the new Ceph backend—BlueStore. It includes optimization recommendations, and MySQL* best-known configurations.
  • Storage
  • An operation guide to build and debug open source Media SDK software stack

    The Goal

    Intel started the Media SDK + associated Linux graphics stack/driver open source project in 2017. Now developers can build the full set of the media software components to run Intel® Media SDK completely from open source repositories with the steps described in this article. Developers can run and debug their media applications with source access to the full stack.

  • Professional
  • Students
  • Linux*
  • Business Client
  • Server
  • C/C++
  • Intermediate
  • Intel® Media Server Studio
  • Intel® Media Server Studio Essentials Edition
  • Intel® Media Server Studio Professional Edition
  • Debugging
  • Development Tools
  • Graphics
  • Media Processing
  • Introduction to the Data Plane Development Kit (DPDK) Eventdev Library

    Learn how to partition the workload for your packet processing application using the DPDK Event Device Library (eventdev). This article provides a technology overview, describes API components, and shows you how to use the eventdev sample app and get statistics.
  • Networking
  • Network Functions Virtualization (NFV)
  • Software Defined Networking (SDN)
  • Tips to Measure the Performance of Matrix Multiplication Using Intel® MKL

    Intel® MKL provide a highly optimized and extensively threaded general matrix-matrix multiplication (GEMM) functions. In this article, we explain how to design and measure of the performance using Intel® MKL SGEMM, and outline about 7 tips to help developers to perform the performance test and quickly evaluate the floating pointing computing capability (FLOPS) on specified processor.
  • Server
  • C/C++
  • Beginner
  • Intel® Math Kernel Library
  • GEMM BLAS matrix multiplication
  • small matrix
  • small problem
  • DGEMM
  • Optimization
  • License Agreement: 

    A Closer Look at Object Detection, Recognition and Tracking

    Learn how object detection, object recognition, and object tracking are distinct processes in computer vision. And also explore some of the algorithms involved with each process.
  • Internet of Things
  • OpenCV*
  • object detection
  • object recognition
  • object tracking
  • Deep Learning
  • Computer Vision
  • Machine Learning
  • Use Intel® QuickAssist Technology Efficiently with NUMA Awareness

    Intel® QuickAssist Technology (Intel® QAT) delivers high-performance encryption and compression. Learn how to configure your Intel QAT driver to take advantage of your system’s NUMA topology.
  • Networking
  • Network Functions Virtualization (NFV)
  • Software Defined Networking (SDN)
  • Subscribe to Technical Article