The Machine Learning Pipeline on AWS

COURSE OUTLINE

Day 1  

Module 0: Introduction 

• Pre-assessment 

Module 1: Introduction to Machine Learning and the ML Pipeline 

• Overview of machine learning, including use cases, types of machine learning, and key concepts 
• Overview of the ML pipeline 
• Introduction to course projects and approach 

Module 2: Introduction to Amazon SageMaker 

• Introduction to Amazon SageMaker 
• Demo: Amazon SageMaker and Jupyter notebooks 
• Hands-on: Amazon SageMaker and Jupyter notebooks 

Module 3: Problem Formulation 

• Overview of problem formulation and deciding if ML is the right solution 
• Converting a business problem into an ML problem 
• Demo: Amazon SageMaker Ground Truth 
• Hands-on: Amazon SageMaker Ground Truth 
• Practice problem formulation 
• Formulate problems for projects 

 

Day 2 Checkpoint 1 and Answer Review 

Module 4: Preprocessing 

• Overview of data collection and integration, and techniques for data preprocessing and visualization 
• Practice preprocessing 
• Preprocess project data 
• Class discussion about projects 

 

Day 3 Checkpoint 2 and Answer Review 

Module 5: Model Training 

• Choosing the right algorithm 
• Formatting and splitting your data for training 
• Loss functions and gradient descent for improving your model
• Demo: Create a training job in Amazon SageMaker 

Module 6: Model Evaluation 

• How to evaluate classification models 
• How to evaluate regression models
• Practice model training and evaluation 
• Train and evaluate project models
• Initial project presentations 

 

Day 4 Checkpoint 3 and Answer Review 

Module 7: Feature Engineering and Model Tuning

• Feature extraction, selection, creation, and transformation
• Hyperparameter tuning 
• Demo: SageMaker hyperparameter optimization 
• Practice feature engineering and model tuning 
• Apply feature engineering and model tuning to projects 
• Final project presentations 

Module 8: Deployment

• How to deploy, inference, and monitor your model on Amazon SageMaker 
• Deploying ML at the edge 
• Demo: Creating an Amazon SageMaker endpoint 
• Post-assessment 
• Course wrap-up