AI Kernel Scheduler - 3.5 English

Vitis AI User Guide (UG1414)

Document ID
Release Date
3.5 English

Real world deep learning applications involve multi-stage data processing pipelines, including many compute-intensive preprocessing operations like data loading from disk, decoding, resizing, color space conversion, scaling, and cropping multiple ML networks like CNN and various post-processing operations like NMS.

The AI kernel scheduler (AKS) application automatically and efficiently pipelines such graphs without much user effort. It provides different kinds of kernels for every stage of the complex graphs, which are highly configurable and plug-and-play. Some examples include preprocessing kernels like image decode and resize, CNN kernels like the Vitis AI DPU kernel and post-processing kernels like SoftMax and NMS. You can create their graphs using kernels and execute their jobs seamlessly for maximum performance.

For more details and examples, see the Vitis AI GitHub (AI Kernel Scheduler).