Machine learning with Azure Databricks (DP-3014)

Azure Databricks is a powerful, cloud-scale platform for data analytics and machine learning. This hands-on class guides you through every step of building, training, optimizing, and deploying machine learning solutions with Azure Databricks. Designed for data scientists and ML engineers, the course covers Apache Spark, MLflow, AutoML, hyperparameter tuning, and deep learning, all within a scalable, collaborative environment. By the end, you’ll have the skills to implement real-world machine learning projects from experimentation to production.

Machine learning with Azure Databricks (DP-3014) Benefits

• In this course, you will learn how to:

  • Learn how to harness the full potential of Databricks for big data analytics and ML.
  • Apply concepts through guided exercises using Spark, MLflow, AutoML, and PyTorch.
  • Gain practical experience managing experiments, tuning hyperparameters, and deploying models at scale.
  • Build expertise in modern machine learning techniques, including deep learning and production-ready workflows.
  • Learn how Unity Catalog and Microsoft Purview support secure, governed, and team-based ML development.

• Training Prerequisites

  To fully benefit from this course, please ensure you possess proficiency in Python for data exploration and machine learning model training using popular open-source frameworks such as Scikit-Learn, PyTorch, and TensorFlow.  

Machine learning with Azure Databricks Training Outline

Learning Objectives

Module 1: Explore Azure Databricks

• Get started with Azure Databricks
• Identify workloads and key concepts
• Understand governance with Unity Catalog & Microsoft Purview
• Hands-on exercise: Explore Azure Databricks

Module 2: Use Apache Spark in Azure Databricks

• Introduction to Spark and clusters
• Work with data files in notebooks
• Visualize data with Spark
• Hands-on exercise: Use Spark in Databricks

Module 3: Train a Machine Learning Model

• Review ML principles
• Prepare data for training
• Train and evaluate models in Databricks
• Hands-on exercise: Train a machine learning model

Module 4: Use MLflow in Azure Databricks

• Manage experiments with MLflow
• Register and serve models
• Hands-on exercise: Track and deploy with MLflow

Module 5: Tune Hyperparameters

• Optimize models with Optuna
• Scale hyperparameter trials
• Hands-on exercise: Hyperparameter tuning in Databricks

Module 6: Use AutoML in Azure Databricks

• Understand AutoML concepts
• Run AutoML experiments via UI and code
• Hands-on exercise: Build models with AutoML

Module 7: Train Deep Learning Models

• Introduction to deep learning concepts
• Train models with PyTorch
• Scale with TorchDistributor
• Hands-on exercise: Train deep learning models

Module 8: Manage ML in Production

• Automate data transformations
• Explore model development and deployment strategies
• Manage versioning and lifecycle
• Hands-on exercise: Deploy and manage ML models