Tutorials

This section contains helpful tutorials that assist users to become familiar with the Qualcomm® AI Engine Direct workflow, and to be able to interact with Qualcomm® AI Engine Direct API and tools to enable their deep learning based use-cases.

Getting Started

• Tutorial: Converting and executing a CNN model with QNN

• Saver Tutorial: Save execution sequence with Saver and replay on a backend

• Sample App Tutorial: Create and build a sample C++ application

Advanced

• QNN HTP Shared Buffer Tutorial

• Executing with DLCs

Custom Operators

• Executing a shallow model using custom op package

• Converting and executing a CNN model with custom operations

Windows

• Windows Hibernation Tutorial

• ARM64X Tutorial

• Windows HTP Library Loading Tutorial

• Windows Error Reporting

Migrating

• Migrating from Hexagon-nn to QNN

LoRA (Low Rank Adaptation)

• LoRA Introduction

• LoRA v2 Overview

• LoRA v2 Offline Flow: Model Preparation

• LoRA v2 Offline Flow: Model Conversion

• LoRA v2 Offline Flow: Generating Binaries

• LoRA v2 Offline Flow: Generating LoRA Binary Sections with QNN

• LoRA v2 Online Flow: QNN Call Flow

• LoRA v2 Online Flow: Genie LoRA API

• LoRA v2 with Graph Switching