Machine learning with Azure Databricks (DP-3014)

Course 8686

  • Duration: 1 day
  • Language: English
  • Level: Intermediate

Embark on an enriching journey with this hands-on instructor-led Microsoft course, 'Machine Learning with Azure Databricks (DP-3014),' designed to empower you with cloud-scale capabilities for data analytics and machine learning. Within this immersive one-day experience, you'll delve into Azure Databricks, a versatile platform enabling data scientists and machine learning engineers to implement robust solutions at scale, revolutionising the way data insights are extracted and utilised.

Machine learning with Azure Databricks Delivery Methods

  • In-Person

  • Online

  • Upskill your whole team by bringing Private Team Training to your facility.

Machine learning with Azure Databricks Training Information

In this course, you will learn how to:

  • Gain proficiency in utilising Azure Databricks, a cloud service offering a scalable platform for data analytics using Apache Spark. 
  • Acquire practical knowledge and hands-on experience in employing Spark to transform, analyse, and visualise data at scale. 
  • Develop skills in training machine learning models and evaluating their performance within the Azure Databricks environment. 
  • Learn to leverage MLflow, an open-source platform for managing the machine learning lifecycle, seamlessly integrated with Azure Databricks. 
  • Master the art of hyperparameter tuning and optimisation using Hyperopt library, enhancing the efficiency of machine learning workflows. 
  • Explore the simplicity and effectiveness of AutoML in Azure Databricks for automating the model building process. 
  • Dive into the realm of deep learning, understanding concepts and training models for complex AI workloads like forecasting, computer vision, and natural language processing. 

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

  • Get started with Azure Databricks 
  • Identify Azure Databricks workloads 
  • Understand key concepts 
  • Exercise Explore Azure Databricks 
  • Knowledge check 
  • Get to know Spark 
  • Create a Spark cluster 
  • Use Spark in notebooks 
  • Use Spark to work with data files 
  • Visualise data 
  • Exercise Use Spark in Azure Databricks 
  • Knowledge check 
  • Understand principles of machine learning 
  • Machine learning in Azure Databricks 
  • Prepare data for machine learning 
  • Train a machine learning model 
  • Evaluate a machine learning model 
  • Exercise Train a machine learning model in Azure Databricks 
  • Knowledge check 
  • Capabilities of MLflow 
  • Run experiments with MLflow 
  • Register and serve models with MLflow 
  • Exercise Use MLflow in Azure Databricks 
  • Knowledge check 
  • Optimise hyperparameters with Hyperopt 
  • Review Hyperopt trials 
  • Scale Hyperopt trials 
  • Exercise Optimise hyperparameters for machine learning in Azure Databricks 
  • Knowledge check 
  • What is AutoML? 
  • Use AutoML in the Azure Databricks user interface 
  • Use code to run an AutoML experiment 
  • Exercise Use AutoML in Azure Databricks 
  • Knowledge check 
  • Understand deep learning concepts 
  • Train models with PyTorch 
  • Distribute PyTorch training with Horovod 
  • Exercise Train deep learning models on Azure Databricks 
  • Knowledge check 

Need Help Finding The Right Training Solution?

Our training advisors are here for you.

Machine learning with Azure Databricks FAQs

Azure Databricks provides a seamless integration of Apache Spark with cloud services, offering a scalable platform tailored for data scientists and machine learning engineers. Its collaborative workspace, optimised Spark performance, and built-in support for machine learning frameworks streamline the process of implementing solutions at scale. 

MLflow simplifies the management of the machine learning lifecycle by providing capabilities for tracking experiments, packaging code, and sharing models. Integrated natively with Azure Databricks, MLflow enables users to efficiently run experiments, register models, and deploy them seamlessly into production environments, facilitating collaboration and reproducibility. 

AutoML in Azure Databricks automates the process of building machine learning models, enabling users to quickly experiment with various algorithms and hyperparameters without extensive manual intervention. By leveraging AutoML, users can accelerate the model development process, optimise model performance, and focus on interpreting and utilising the generated insights. 

Azure Databricks provides a comprehensive environment for training deep learning models, including support for popular frameworks like PyTorch. With capabilities for distributed training using tools like Horovod, users can efficiently scale their deep learning workflows to handle large datasets and complex AI workloads, empowering them to unlock new possibilities in areas such as computer vision and natural language processing.