DeepLearning.AI

Introduction to On-Device AI

up to 1 hour
Beginner

Learn to deploy AI models on edge devices like smartphones, using their local compute power for faster and more secure inference. This course covers model conversion, quantization, and device integration, equipping you with the skills to optimize and deploy AI models on billions of devices.

Model conversion
Quantization
Device integration
Neural network graph capture
On-device compilation

Overview

This course provides comprehensive training on deploying AI models on edge devices, reducing latency, enhancing efficiency, and preserving privacy. You will learn key concepts such as neural network graph capture, on-device compilation, and hardware acceleration. The course includes practical exercises like converting pretrained models from PyTorch and TensorFlow, deploying real-time image segmentation models, and integrating models into Android apps. By the end of the course, you will be equipped to develop and deploy AI models on billions of devices, optimizing them for efficient on-device performance.

Online
course location
English
course language
Self-paced
course format
Live classes
delivered online

Who is this course for?

Beginner AI Developers

Individuals new to AI development looking to understand and deploy AI models on edge devices.

ML Engineers

Machine Learning engineers aiming to optimize and deploy models on devices like smartphones and IoT devices.

Mobile Developers

Developers focused on mobile applications who want to integrate AI models into their apps for enhanced performance and capabilities.

This course equips you with the skills to deploy AI models on edge devices, enhancing efficiency and preserving privacy. Ideal for beginner AI developers, ML engineers, and mobile developers, it covers key concepts and practical exercises to help you optimize and deploy AI models on billions of devices.

Pre-Requisites

1 / 2

Familiarity with Python
Basic knowledge of PyTorch or TensorFlow

What will you learn?

Introduction to On-Device AI

Learn to deploy AI models on edge devices like smartphones, using their local compute power for faster and more secure inference.

Model Conversion

Explore model conversion by converting your PyTorch/TensorFlow models for device compatibility, and quantize them to achieve performance gains while reducing model size.

Device Integration

Learn about device integration, including runtime dependencies, and how GPU, NPU, and CPU compute unit utilization affect performance.

On-Device Deployment

Explore how deploying models on device reduces latency, enhances efficiency, and preserves privacy.

Neural Network Graph Capture

Go through key concepts of on-device deployment such as neural network graph capture, on-device compilation, and hardware acceleration.

Real-Time Image Segmentation

Deploy a real-time image segmentation model on device with just a few lines of code.

Model Performance Testing

Test your model performance and validate numerical accuracy when deploying to on-device environments.

Quantization

Quantize and make your model up to 4x faster and 4x smaller for higher on-device performance.

Android App Integration

See a demonstration of the steps for integrating the model into a functioning Android app.

Meet your instructor

Krishna Sridhar
Engineering Leader, Qualcomm
Krishna Sridhar is an experienced engineering leader passionate about building wonderful products powered by machine learning. He currently serves as the Senior Director of Engineering at Qualcomm, where he leads the development of the Qualcomm® AI Hub.

Upcoming cohorts

Cost
Free
Duration
1 hour
Dates
start now
Location
Online

Free