MIVisionX v1.9.1

Release Notes

Online Documentation

MIVisionX Documentation

Features

• MIVisionX Setup Updates
• Support with MIOpen V2.5.0 with OpenCL & MIOpenGEMM
• Minor Bug-fixes
• Cleanup

NOTE:

• Run Setup Script - MIVisionX-setup.py - V1.7.13 with --reinstall option if dependencies are previously installed

Release code checkout

git clone -b 1.9.1 https://github.com/GPUOpen-ProfessionalCompute-Libraries/MIVisionX

Prerequisites

Linux

• Ubuntu 16.04/18.04/20.04 or CentOS 7.5/7.6/8.1
• ROCm supported hardware
• ROCm
• Run Setup Script - MIVisionX-setup.py - V1.7.13

Windows without Windows Machine Learning Module

• Windows 10
• Windows SDK
• Visual Studio 2017
• Install OpenCL SDK
• OpenCV 3.4
  • Set OpenCV_DIR environment variable to OpenCV/build folder
  • Add %OpenCV_DIR%\x64\vc14\bin or %OpenCV_DIR%\x64\vc15\bin to your PATH

Windows with Windows Machine Learning Module

• Windows 10, version 1809 or later
• Windows SDK, build 17763 or later
• Visual Studio 2017, version 15.7.4 or later
  • Visual Studio extension for C++/WinRT
• Install OpenCL SDK
• OpenCV 3.4
  • Set OpenCV_DIR environment variable to OpenCV/build folder
  • Add %OpenCV_DIR%\x64\vc14\bin or %OpenCV_DIR%\x64\vc15\bin to your PATH

Known Issues

• Package (.deb & .rpm) install requires OpenCV v3.4.0 to execute AMD OpenCV extensions
• ROCm 3.0 and above has known to slow down OpenCL kernels.

Tested configurations

• Windows 10
• Linux: Ubuntu - 16.04/18.04/20.04 & CentOS - 7.5/7.6/8.1
• ROCm: rocm-dkms - 3.3.0-19 & 3.6.0-43
• rocm-cmake - github master:ac45c6e
• MIOpenGEMM - 1.1.5
• MIOpen - 2.5.0
• Protobuf - V3.12.0
• OpenCV - 3.4.0
• RPP - 0.4
• Dependencies for all the above packages
• MIVisionX Setup Script - V1.7.13