DP-3014 Build machine learning solutions using Azure Databricks

Course DP-3014 Build machine learning solutions using Azure Databricks

Azure Databricks is a cloud-scalable platform for data analytics and machine learning. Data scientists and machine learning engineers can use Azure Databricks to deploy machine learning solutions at scale.

Level: Intermediate - Role: Data Scientist - Product: Azure Databricks

Objectives of the DP-3014 training

• Explore how Azure Databricks enables data scientists and machine learning engineers to implement machine learning solutions at scale.
• Leverage the cloud scaling capabilities of Azure Databricks to train and deploy machine learning models.
• Use open source tools and frameworks, such as Scikit-Learn, PyTorch, and TensorFlow, in the Azure Databricks environment.
• Use open source tools and frameworks, such as Scikit-Learn, PyTorch, and TensorFlow, in the Azure Databricks environment.
• Optimize and manage Machine Learning models for advanced data analysis and effective predictions.
• Develop practical experience in implementing machine learning workflows in an Azure Databricks environment.

Course content DP-3014: Create machine learning solutions with Azure Databricks

Module 1: Explore Azure Databricks

• Introduction to Azure Databricks
• Identifying Azure Databricks workloads
• Description of key concepts
• Exercise: Explore Azure Databricks

Module 2: Using Apache Spark in Azure Databricks

• Discover Spark
• Creating a Spark cluster
• Using Spark in notebooks
• Using Spark to work with data files
• Data visualization
• Exercise: Using Spark in Azure Databricks

Module 3: Training a Machine Learning model in Azure Databricks

• Description of the principles of machine learning
• Machine learning in Azure Databricks
• Preparing data for machine learning
• Training a Machine Learning model
• Evaluate a Machine Learning model
• Exercise: Training a Machine Learning model in Azure Databricks

Module 4: Using MLflow in Azure Databricks

• MLflow functionalities
• Running experiments with MLflow
• Registration and service of models with MLflow
• Exercise: Using MLflow in Azure Databricks

Module 5: Hyperparameter Tuning in Azure Databricks

• Hyperparameter optimization with Hyperopt
• Hyperopt Testing Review
• Hyperopt Test Scale
• Exercise: Hyperparameter optimization for machine learning in Azure Databricks

Module 6: Using AutoML in Azure Databricks

• What is AutoML?
• Using AutoML in the Azure Databricks user interface
• Using code to run an AutoML experiment
• Exercise: Using AutoML in Azure Databricks

Module 7: Training deep learning models in Azure Databricks

• Understanding the concepts of deep learning
• Model training with PyTorch
• PyTorch training distribution with Horovod
• Exercise: Training deep learning models in Azure Databricks

Module 8: Manage machine learning in production with Azure Databricks

• Introduction
  
• Automate your data transformations
  
• Explore the development of models
  
• Explore model implementation strategies
  
• Explore model version control and lifecycle management
  
• Exercise: Managing a machine learning model

Prerequisites

Experience using Python to explore data and train Machine Learning models with common open source frameworks such as Scikit-Learn, PyTorch, and TensorFlow is recommended.

Language

• Course: English / Spanish
• Labs: English / Spanish