Imagination helps developers of CNNs with neural network software dev kit
Imagination Technologies announces the first PowerVR CLDNN software development kit (SDK) for developing neural network applications on PowerVR graphics processor units (GPUs). The neural network SDK makes it easy for developers to create convolutional neural networks (CNNs) using PowerVR hardware, says Imagination. Both an application processor interface (API) and SDK are included, as well as an image for flashing on to an Acer Chromebook R13 for hardware development.
Imagination launched its first neural network accelerator, PowerVR Series2NX in last year. Future versions of the API and SDK library will enable developers to develop neural network applications that will be able to target Imagination’s GPU and neural network accelerator (NNA) hardware without having to rewrite code.
Testing and deploying accelerated neural networks is increasing as automotive OEMs adopt advanced driver assistance systems (ADAS) and autonomous driving. At the same time, smartphone users expect technologies such as facial recognition, and smart home technologies need to enhance security the user experience by recognising objects and people.
The PowerVR CLDNN API is Imagination’s first artificial intelligence (AI) -focused API. It provides functions to create network layers for constructing and running a neural network on PowerVR hardware. By using specialist OpenCL kernels, it enables developers to focus on their neural network creation with fewer overheads. The API also performs low-level hardware-specific optimisations, enabling better-optimised graphs than a custom user OpenCL implementation.
The PowerVR CLDNN SDK demonstrates how a neural network can be deployed to PowerVR hardware through the CLDNN API. It includes helper functions such as file loading, dynamic library initialisation and OpenCL content management. A reference manual is also included, as well as the source code for sample applications that demonstrate how to use the PowerVR CLDNN API.
An image has also been created that developers can flash to an Acer Chromebook R13, which contains a PowerVR GX6250 GPU via the Mediatek MT-8173 SoC. By downloading the image, developers can access everything they need to write neural network applications on the device, says Imagination Technologies.