Releases: ROCm/MIVisionX
MIVisionX for ROCm 3.8.0
MIVisionX for ROCm 3.7.0
MIVisionX v1.9.1
Release Notes
Online 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 CentOS7.5
/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 toOpenCV/build
folder - Add
%OpenCV_DIR%\x64\vc14\bin
or%OpenCV_DIR%\x64\vc15\bin
to yourPATH
- Set
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 toOpenCV/build
folder - Add
%OpenCV_DIR%\x64\vc14\bin
or%OpenCV_DIR%\x64\vc15\bin
to yourPATH
- Set
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
MIVisionX v1.9
Release Notes
Online Documentation
NEW Features
- AMD RPP Extension - New Features & Bug fix
- RALI - New Features & Bug-fixes
- Minor Bug-fixes
- ADAT - Bug Fixes
- Cleanup
Release code checkout
git clone -b 1.9 https://github.com/GPUOpen-ProfessionalCompute-Libraries/MIVisionX
Prerequisites
Linux
- Ubuntu
16.04
/18.04
/20.04
or CentOS7.5
/7.6
/8.1
- ROCm supported hardware
- ROCm
- Run Setup Script -
MIVisionX-setup.py
-V1.7.11
Windows without Windows Machine Learning Module
- Windows 10
- Windows SDK
- Visual Studio 2017
- Install OpenCL SDK
- OpenCV 3.4
- Set
OpenCV_DIR
environment variable toOpenCV/build
folder - Add
%OpenCV_DIR%\x64\vc14\bin
or%OpenCV_DIR%\x64\vc15\bin
to yourPATH
- Set
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 toOpenCV/build
folder - Add
%OpenCV_DIR%\x64\vc14\bin
or%OpenCV_DIR%\x64\vc15\bin
to yourPATH
- Set
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
MIVisionX Lite V1.0
Release Notes
Online Documentation
Features
- OpenVX 1.0.1
- OpenVX - OpenCV Extension
- RunCL
- RunVX
Release code checkout
git clone -b 1.0 https://github.com/GPUOpen-ProfessionalCompute-Libraries/MIVisionX
Prerequisites
Linux
- Ubuntu
16.04
/18.04
or CentOS7.5
/7.6
- ROCm supported hardware
- ROCm
- Run Setup Script -
MIVisionX-Lite-setup.py
MacOS
- Install Homebrew
- CMake -
brew install cmake
- Git -
brew install git
- Run Setup Script -
MIVisionX-Lite-setup.py
Windows
- Windows 10
- Windows SDK
- Visual Studio 2017
- Install OpenCL SDK
- OpenCV 3.4
- Set
OpenCV_DIR
environment variable toOpenCV/build
folder - Add
%OpenCV_DIR%\x64\vc14\bin
or%OpenCV_DIR%\x64\vc15\bin
to yourPATH
- Set
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.
- If OpenCL failure occurs on macOS, set environment variable to run on CPU by default
export AGO_DEFAULT_TARGET=CPU
Tested configurations
- Windows 10
- Linux: Ubuntu -
16.04
/18.04
& CentOS -7.5
/7.6
- ROCm: rocm-dkms -
3.3.0-19
- OpenCV - 3.4.0
- Dependencies for all the above packages
MIVisionX v1.8
Release Notes
Online Documentation
NEW Features
- New AMD RPP Extension features
- RALI - Bug-fixes
- Minor Bug-fixes
- ADAT - Bug Fixes
- Cleanup
Release code checkout
git clone -b 1.8 https://github.com/GPUOpen-ProfessionalCompute-Libraries/MIVisionX
Prerequisites
Linux
- Ubuntu
16.04
/18.04
or CentOS7.5
/7.6
- ROCm supported hardware
- ROCm
- Run Setup Script -
MIVisionX-setup.py
Windows without Windows Machine Learning Module
- Windows 10
- Windows SDK
- Visual Studio 2017
- Install OpenCL SDK
- OpenCV 3.4
- Set
OpenCV_DIR
environment variable toOpenCV/build
folder - Add
%OpenCV_DIR%\x64\vc14\bin
or%OpenCV_DIR%\x64\vc15\bin
to yourPATH
- Set
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 toOpenCV/build
folder - Add
%OpenCV_DIR%\x64\vc14\bin
or%OpenCV_DIR%\x64\vc15\bin
to yourPATH
- Set
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
MIVisionX v1.7
Release Notes
Online Documentation
NEW Features
- New Applications & Samples
- RALI - Bug-fixes
- Minor Bug-fixes
- ROCm 3.1 support
Release code checkout
git clone -b 1.7 https://github.com/GPUOpen-ProfessionalCompute-Libraries/MIVisionX
Prerequisites
Linux
- Ubuntu
16.04
/18.04
or CentOS7.5
/7.6
- ROCm supported hardware
- ROCm
- Run Setup Script -
MIVisionX-setup.py
Windows without Windows Machine Learning Module
- Windows 10
- Windows SDK
- Visual Studio 2017
- Install OpenCL SDK
- OpenCV 3.4
- Set
OpenCV_DIR
environment variable toOpenCV/build
folder - Add
%OpenCV_DIR%\x64\vc14\bin
or%OpenCV_DIR%\x64\vc15\bin
to yourPATH
- Set
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 toOpenCV/build
folder - Add
%OpenCV_DIR%\x64\vc14\bin
or%OpenCV_DIR%\x64\vc15\bin
to yourPATH
- Set
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
& CentOS -7.5
/7.6
- ROCm: rocm-dkms -
3.1.44
- rocm-cmake - github master:ac45c6e
- MIOpenGEMM - 1.1.5
- MIOpen - 2.1.0
- Protobuf - V3.5.2
- OpenCV - 3.4.0
- Dependencies for all the above packages
MIVisionX v1.6
Release Notes
Online Documentation
NEW Features
- New Applications & Samples
- RALI - Bug-fixes
- Minor Bug-fixes
- ROCm 2.10 support
- Duplicate data cleanup
- Package optimized
Release code checkout
git clone -b 1.6 https://github.com/GPUOpen-ProfessionalCompute-Libraries/MIVisionX
Install Packages on Linux
.deb
sudo dpkg -i mivisionx-1.6-1.x86_64.deb
sudo apt-get install -f
.rpm
sudo yum install mivisionx-1.6-1.x86_64.rpm
NOTE: Prerequisites for apt-get
/yum install
- Ubuntu
16.04
/18.04
or CentOS7.5
/7.6
- ROCm supported hardware
- ROCm
Install Packages on Windows
MIVisionX-installer.msi
- Prerequisites
- Windows 10
- Windows SDK
- Visual Studio 2017
- Install OpenCL SDK
- OpenCV 3.4
- Set
OpenCV_DIR
environment variable toOpenCV/build
folder - Add
%OpenCV_DIR%\x64\vc14\bin
or%OpenCV_DIR%\x64\vc15\bin
to yourPATH
- Set
MIVisionX_WinML-installer.msi
- Prerequisites
- 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 toOpenCV/build
folder - Add
%OpenCV_DIR%\x64\vc14\bin
or%OpenCV_DIR%\x64\vc15\bin
to yourPATH
- Set
Known Issues
- Package (.deb & .rpm) install requires OpenCV
v3.4.0
to execute AMD OpenCV extensions
Tested configurations
- Windows 10
- Linux: Ubuntu -
16.04
/18.04
& CentOS -7.5
/7.6
- ROCm: rocm-dkms -
2.10.14
- rocm-cmake - github master:ac45c6e
- MIOpenGEMM - 1.1.5
- MIOpen - 2.1.0
- Protobuf - V3.5.2
- OpenCV - 3.4.0
- Dependencies for all the above packages
MIVisionX v1.5
Release Notes
Online Documentation
NEW Features
- RALI - Bug-fixes
- Minor Bug-fixes
- ROCm 2.9 support
- SLES Build support
Release code checkout
git clone -b 1.5 https://github.com/GPUOpen-ProfessionalCompute-Libraries/MIVisionX
Install Packages on Linux
.deb
sudo dpkg -i mivisionx-1.5-1.x86_64.deb
sudo apt-get install -f
.rpm
sudo yum install mivisionx-1.5-1.x86_64.rpm
NOTE: Prerequisites for apt-get
/yum install
- Ubuntu
16.04
/18.04
or CentOS7.5
/7.6
- ROCm supported hardware
- ROCm
Install Packages on Windows
MIVisionX-installer.msi
- Prerequisites
- Windows 10
- Windows SDK
- Visual Studio 2017
- Install OpenCL SDK
- OpenCV 3.4
- Set
OpenCV_DIR
environment variable toOpenCV/build
folder - Add
%OpenCV_DIR%\x64\vc14\bin
or%OpenCV_DIR%\x64\vc15\bin
to yourPATH
- Set
MIVisionX_WinML-installer.msi
- Prerequisites
- 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 toOpenCV/build
folder - Add
%OpenCV_DIR%\x64\vc14\bin
or%OpenCV_DIR%\x64\vc15\bin
to yourPATH
- Set
Known Issues
- Package (.deb & .rpm) install requires OpenCV
v3.4.0
to execute AMD OpenCV extensions
Tested configurations
- Windows 10
- Linux: Ubuntu -
16.04
/18.04
& CentOS -7.5
/7.6
- ROCm: rocm-dkms -
2.9.6
- rocm-cmake - github master:ac45c6e
- MIOpenGEMM - 1.1.5
- MIOpen - 2.1.0
- Protobuf - V3.5.2
- OpenCV - 3.4.0
- Dependencies for all the above packages
MIVisionX v1.4
Release Notes
Online Documentation
NEW Features
- RALI - Radeon Augmentation Library is designed to efficiently decode and process images and videos from a variety of storage formats and modify them through a processing graph programmable by the user.
- Extended NNEF Support
- Neural Net Model Compiler & Optimizer updates
- Extended ONNX support
- Minor Bug Fixes
Release code checkout
git clone -b 1.4 https://github.com/GPUOpen-ProfessionalCompute-Libraries/MIVisionX
Install Packages on Linux
.deb
sudo dpkg -i mivisionx-1.4-1.x86_64.deb
sudo apt-get install -f
.rpm
sudo yum install mivisionx-1.4-1.x86_64.rpm
NOTE: Prerequisites for apt-get
/yum install
- Ubuntu
16.04
/18.04
or CentOS7.5
/7.6
- ROCm supported hardware
- ROCm
Install Packages on Windows
MIVisionX-installer.msi
- Prerequisites
- Windows 10
- Windows SDK
- Visual Studio 2017
- Install OpenCL SDK
- OpenCV 3.4
- Set
OpenCV_DIR
environment variable toOpenCV/build
folder - Add
%OpenCV_DIR%\x64\vc14\bin
or%OpenCV_DIR%\x64\vc15\bin
to yourPATH
- Set
MIVisionX_WinML-installer.msi
- Prerequisites
- 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 toOpenCV/build
folder - Add
%OpenCV_DIR%\x64\vc14\bin
or%OpenCV_DIR%\x64\vc15\bin
to yourPATH
- Set
Known Issues
- Package (.deb & .rpm) install requires OpenCV
v3.4.0
to execute AMD OpenCV extensions - Latest MIOpen versions with OpenCL backend has linking errors with MIOpenGEMM. If you are facing problems with MIOpen revert back to
1.8.1
or rerun the MIVisionX-setup script. - ROCm
2.8
and above is know to stall MIVisionX Neural Net Workflow, use ROCm2.7.22
with MIVisionX till the issue is solved. Use MIVisionX Dockers with ROCm2.7.22
if you cannot downgrade ROCm on your machine.
Tested configurations
- Windows 10
- Linux: Ubuntu -
16.04
/18.04
& CentOS -7.5
/7.6
- ROCm: rocm-dkms -
2.7.22
- rocm-cmake - github master:ac45c6e
- MIOpenGEMM - 1.1.5
- MIOpen - 1.8.1
- Protobuf - V3.5.2
- OpenCV - 3.4.0
- Dependencies for all the above packages