Profiling AI models using Qualcomm AI Engine Direct

Unlock the full potential of your AI models on Qualcomm® hardware! This video walks developers through profiling AI models using the Qualcomm® AI Engine Direct SDK, covering essential steps and best practices for optimising performance on Qualcomm® Kryo™ CPU, Adreno™ GPU, and Hexagon™ NPU. Chapters include: Prerequisites before profiling Qualcomm AI Engine Direct SDK workflow qnn-net-run Qualcomm Neural Network linting profiling Model optimisation example using linting profiling Identifying the bottleneck Additional guidelines Learn how to target specific accelerators, delegate workloads from TensorFlow Lite or ONNX runtime, and use unified APIs to get closer to the silicon for maximum efficiency. Find out more about the SDK: https://www.qualcomm.com/developer/software/qualcomm-ai-engine-direct-sdk If you have any questions or need further assistance, feel free to leave a comment below. Subscribe: More Qualcomm Developer videos: http://tinyurl.com/2p8xmcw6 Join our community of developers on Discord to stay up-to-date, find expert support, and access exclusive virtual events: https://discord.gg/THUPBtskgs