Machine Learning with Amazon SageMaker

With the knowledge you gain from this course, you can quickly develop Machine Learning solutions on Amazon Web Services (AWS).

You will be introduced to techniques for data exploration to help you find patterns, meaning, and value in your company data through hands-on laboratories and practical activities.

At the end of the course, participants will be able to:

  • Understand Machine Learning (ML) concepts
  • Prepare your data
  • Upload your dataset to Amazon SageMaker
  • Train, analyze, and tune an ML model
  • Analyze and publish your results 

Anyone who needs to develop machine learning solutions to analyze data and wants to take use of Amazon's cloud-based data science platform.

  • Overview of key Machine Learning and Deep Learning concepts
  • Getting about in AWS
  • Overview of SageMaker features
  • Taking your first look at SageMaker studio
  • Identifying your data and articulating your problem
  • Format data for consistency
  • Cleaning and validating your data
  • Uploading to SageMaker

  • Clustering
  • Trend analysis
  • Finding other relationships between different types of data
  • Frequency tables
  • Cross-tabulation tables
  • Bar charts
  • Line graphs
  • Pie charts
  • Heat Maps
  • Scatter graphs

  • Creating a Training job
  • Assigning Compute resources
  • Selecting the right algorithm
  • Overview of using custom code (Python, TensorFlow)
  • SageMaker Hosting Services
  • Configuring and creating an HTTPS endpoint

  • Making inferences from your dataset
  • Indexing and real-time indices
  • Using Batch Transform to preprocess data to train a new model
  • SageMaker Debugger
  • Offline testing
  • Online testing
  • Validating using a holdout set

  • Defining metrics
  • Hyperparameter tuning
  • Automatic model tuning
  • Creating, Storing and Sharing Features
  • Online / Offline
  • Feature Groups
  • Discovery
  • Batch Inference
  • Feature Data Ingestion

Related Courses