ML Introduction & Data Preparation

4 hours of instruction

This course kickstarts the series of courses on MLOps technical implementation. It lays the ground for MLOps terminology and lets the students dive into the initial preparatory stages of setting up cloud platform services required to prepare data for the next stages of the ML cycle in production environment.

OBJECTIVES

1. Set up and configuration of services required to perform data preparation
2. Data ingestion, exploring and validation

PREREQUISITES

Introduction to MLOps Theory

SYLLABUS & TOPICS COVERED

1. Fundamentals of AWS SageMaker
   • ‘Discuss AWS SageMaker and how it is used for MLOps’
   • ‘Set up SageMaker domain’
   • ‘Identify and locate SageMaker Studio features’
2. Setting up an MLOps project
   • ‘Identify the integrations in SageMaker Studio’
   • ‘Summarize the features and uses of the MLOps template’
   • ‘Set up the workspace’
3. Interpreting SageMaker’s file repositories
   • ‘Describe the file structure of ModelBuild Repo’
   • ‘Describe the file structure of ModelDeploy Repo’
   • ‘Summarize the files and functionality of ModelMonitor Repo’
4. Preparing a dataset
   • ‘Summarize the Heart Failure Prediction Dataset’
   • ‘Complete the setup to pull the dataset to the S3 Bucket’
   • ‘Set the dataset S3 path as parameter to the pipeline’

SOFTWARE REQUIREMENTS

API Gateway, AWS Sagemaker, Access to AWS accounts, CodeBuild, CodePipeline, Lambda, S3