QUALCOMM HAS OPENED UP its Snapdragon-based deep learning software development kit (SDK) to all.
Qualcomm's Neural Processing Engine was first announced in May last year with the aim of making it easier for developers to run deep learning programmes on devices that use its Snapdragon processors, such as mobiles, tablets, and even drones.
For example, if you're a developer building an app that uses AI for face recognition or object tracking, you can integrate Qualcomm's SDK and it will run faster on phones with compatible processors.
On Tuesday, the firm announced that, for the first time, the SDK is now available to all via the Qualcomm Developer Network.
In a blog post announcing the first public SDK, Qualcomm boasts that Facebook has already begun using Snapdragon NPE, and plans to integrate it into the camera of the Facebook app to accelerate Caffe2-powered AR features.
According to the lawsuit-laden chip firm, Facebook is seeing improvements in performance for its AR features on images and live video when using Qualcomm's Adreno GPUs on Snapdragon SoCs.
"We developed the Snapdragon NPE SDK to help meet the increasing demand for running deep neural networks efficiently on mobile and other edge devices powered by Snapdragon," said Gary Brotman, director of product management at Qualcomm.
"The Snapdragon NPE was created to give developers the tools to easily migrate intelligence from the cloud to edge devices, and affords them the flexibility to target the appropriate Snapdragon core with the power and performance profile for a given user experience."
The Snapdragon NPE is compatible with Qualcomm's Snapdragon 600 and 800 Series processors, and supports a number of deep learning frameworks including Caffe2 and Tensorflow.
Qualcomm notes that the SDK includes runtime software, libraries, APIs, offline model conversion tools, sample code, documentation, and debugging and benchmarking tools, and adds that devs Developers can choose the optimal Snapdragon core for the desired user experience, including the Qualcomm Kryo CPU, Qualcomm Adreno GPU or the Qualcomm Hexagon DSP. µ
Though it's not exactly an even playing field
Biz machines offer of Intel CPUs and AMD or Nvidia graphics
Thermal throttling is hampering the overclockable chip's performance
Firm is making its Lightning port even more locked down