Use Amazon SageMaker Canvas to make your first ML Model

Languages Available: Español (Latinoamérica) | Français | Bahasa Indonesia | 日本語 | 한국어 | Português (Brasil) | 中文（简体)

This lab demonstrates how to use Amazon SageMaker Canvas to create a machine learning (ML) model to determine whether a customer is retained based on an email campaign for new products and services.
You are a business analyst on a marketing team for a chain of retail stores. The team has lots of data on whether customers are retained after an initial purchase based on various email marketing campaigns for new and existing product and service offerings. The team is trying to predict if these marketing campaigns will retain customers’ interest in the products and services the company offers.
Your chief marketing officer wants you to use the data to conduct a proof of concept to make predictions on the effectiveness of these campaigns. You have contacted the company’s IT team and they have recommended you use SageMaker Canvas to explore the data and make predictions, because it does not require deep ML expertise through a data scientist to make predictions.

Level

Fundamental

Duration

1 Hours 15 Minutes

Course Objectives

In this course, you will learn how to:

• Import data from Amazon Simple Storage Service (Amazon S3) into SageMaker Canvas.
• Use SageMaker Canvas to manage the data, complete feature engineering tasks, and select a feature to predict.
• Use SageMaker Canvas to build and train a model to predict whether a customer is retained.
• Review the lab architecture and security implementation.
• Use SageMaker Canvas to evaluate the model.
• Make predictions using sample data.

Intended Audience

This course is intended for:

Data Scientists responsible for using large data sets to address business issues Deploy sophisticated analytics programs, machine learning and statistical methods.

Prerequisites

We recommend that attendees of this course have the following prerequisites:

• Access to a computer with Microsoft Windows, macOS, or Linux (Ubuntu, SuSE, or Red Hat)
• A modern internet browser such as Chrome or Firefox

Course Outline

Task 1: Access SageMaker Canvas and review S3 buckets
Task 2: Create training and validation datasets using AWS Cloud9 and a Python script
Task 3: Browse S3 and import the training dataset into SageMaker Canvas
Task 4: Choose a target for prediction
Task 5: Review the model recipe and prepare and analyze the data
Task 6: Train the model
Task 7: Make predictions with the validation dataset
Task 8: Retrieve model artifacts