More info
Full Description
Scope
This standard defines a set of application programming interfaces (APIs) that can be used on different deep learning inference engines. The interfaces include parameter reading, model compilation optimization, operator registration, thread management, input/output data acquisition, inference instance creation and inference execution.
Purpose
This standard defines efficient APIs for model management, device management, input/output, and inference which break down the barriers between different deep learning inference engines and applications. The target deep learning applications include natural language processing, image classification, target detection.
Abstract
New IEEE Standard - Active - Draft.A set of application program interfaces (APIs) is defined in this standard, which is aimed at breaking down the barriers between different deep learning inference engines and applications. The APIs are comprised of functional interfaces including parameter reading, model compilation optimization, operator registration, thread management, input/output data acquisition, inference instance creation and inference execution.