Using SageMaker Notebooks to Train and Deploy Machine Learning Models

Amazon SageMaker notebooks provide a fully-managed environment for machine learning and data science development. You will use a SageMaker notebook instance to train and deploy a machine learning model using Python. You will go through the process of preparing raw data for use with machine learning algorithms. Then you will use a built-in SageMaker algorithm to train a model using the prepared data. Lastly, you will use SageMaker to host the trained model and learn how you can make real-time predictions using the model.

Lab Objectives

Upon completion of this Lab you will be able to:

• Use SageMaker notebook instances to run Jupyter Notebooks
• Write code using the Python Data Analysis Library (pandas) and the SageMaker Python SDK to:
  • train models using built-in SageMaker algorithms
  • Create SageMaker models
  • Deploy SageMaker endpoints to get real-time inferences from your models

Intended Audience

This lab is intended for:

• Anyone interested in using SageMaker to build and deploy machine learning models in code

Prerequisites

You should be familiar with:

• Some knowledge of machine learning concepts is beneficial, but not required
• Basic programming using Python 3
• Completion of the Forecast Flight Delays with Amazon SageMaker lab is recommended for a deeper understanding of the data used in this lab 
• Basic S3 concepts

Updates

September 7th, 2023 - Resolved an issue installing package dependencies in the lab notebook

June 17th, 2023 - Resolved training job issue

January 10th, 2022 - Updated notebook to ensure dependencies are up to date

December 2nd, 2020 - Updated code to be compliant with the SageMaker v2 library; Modified code to prevent training job name collisions