Deep Learning on AWS

Intended audience

This course is intended for:

• Developers who are responsible for developing deep learning applications
• Developers who want to understand the concepts behind deep learning and how to implement a deep learning solution on AWS Cloud

Prerequisites

We recommend that attendees of this course have a basic understanding of:

• ML processes
• AWS core services like Amazon EC2 and knowledge of AWS SDK
• A scripting language like Python

Day One

Module 1: Machine learning overview

• A brief history of AI, ML, and DL
• The business importance of ML
• Common challenges in ML
• Different types of ML problems and tasks
• AI on AWS

Module 2: Introduction to deep learning

• Introduction to DL
• The DL concepts
• A summary of how to train DL models on AWS
• Introduction to Amazon SageMaker
• Hands-on lab: Spinning up an Amazon SageMaker notebook instance and running a multi-layer perceptron neural network model

Module 3: Introduction to Apache MXNet

• The motivation for and benefits of using MXNet and Gluon
• Important terms and APIs used in MXNet
• Convolutional neural networks (CNN) architecture
• Hands-on lab: Training a CNN on a CIFAR-10 dataset

Module 4: ML and DL architectures on AWS

• AWS services for deploying DL models (AWS Lambda, AWS IoT Greengrass, Amazon ECS, AWS Elastic Beanstalk)
• Introduction to AWS AI services that are based on DL (Amazon Polly, Amazon Lex, Amazon Rekognition)
• Hands-on lab: Deploying a trained model for prediction on AWS Lambda