tagged news

The Khronos Group NNEF Working Group Chair Peter McGuinness discusses fragmentation in the Machine Learning field. Machine learning capabilities are being added to everything from social media platforms, IoT devices and cameras to robots and cars. The pace of innovation is leading to fragmentation, and one potential consequence of that fragmentation is a risk of stalling. A universal transfer standard for neural networks will cut down time wasted on transfer and translation and provide a comprehensive, extensible and well-supported solution that all parts of the ecosystem can depend on. The Neural Network Exchange Format is one of two standards currently being developed to satisfy this need. Learn more about NNEF and how it aims to solve this issue.

LunarG has updated its white paper that describes using spirv-opt to reduce SPIR-V size. New discussions include using spirv-opt to “legalize” SPIR-V when generated from HLSL; SPIR-V integration with the glslang and dxc frontends; description of new size reduction passes available in spirv-opt; and updates to the recommended recipe for those who wish to customize their optimization.

The X-Plane cross-platform flight simulator has been depending upon OpenGL for nearly two decades since the program first came into existence, but a port of its rendering engine to use the Vulkan API has been a work-in-progress. It looks like their Vulkan support is getting squared away as the company has tweeted this weekend they will be talking about Vulkan integration this weekend at the Flight Sim Expo in Las Vegas.

Qualcomm Technologies debuted the Qualcomm Snapdragon XR1 Platform, a dedicated Extended Reality (XR) platform during a launch event leading up to the Augmented World Expo (AWE). The integrated display processor provides a range of display options with hardware accelerated composition, dual-display support, 3D overlays and support for leading graphics Application Programming Interfaces (API), including OpenGL, OpenCL and Vulkan.

Synopsys, Inc. released OpenVX 1.2 conformant DesignWare ARC MetaWare EV Development ToolkitSynopsys, Inc. announced a new release of its DesignWare ARC MetaWare EV Development Toolkit. The ARC MetaWare EV Development Toolkit offers a programming environment for the EV6x vision processor IP based on the OpenVX open standard API, with a C/C++ compiler and OpenCL C vectorizing compiler. The Toolkit includes OpenVX kernels optimized to run on EV6x Processors, the first hardware-software platform to successfully pass The Khronos Group’s OpenVX 1.2 conformance tests.

The Khronos Group last week announced the launch of a project with Au-zone of Calgary, Canada to produce two conversion tools that will allow developers to import NNEF files into TensorFlow and Caffe2 as well as export NNEF files from those training frameworks. In line with Khronos’ recent policy, the tools will be made available in the second half of 2018 as open source projects on Github where contributions from the open source community will be welcomed. These tools will join the converters for Caffe and Tensorflow (which has two formats to consider) and the NNEF parser that are all already available.

The Khronos™ Group is working with Au-Zone Technologies to enable NNEF™ (Neural Network Exchange Format) files to be easily used with leading machine learning training frameworks. NNEF enables the optimized ingestion of trained neural networks into hardware inference engines on a diverse range of devices and platforms. Au-Zone is working with the Khronos NNEF Working Group to implement two purpose-built bidirectional converters, between TensorFlow and NNEF and also Caffe2 and NNEF. Both converters are expected to be released as open source projects to the development community in Q3 2018 under the Apache 2.0 license.