- Module 1: Get started with Azure Databricks
After completing this module, you will be able to:
- Understand Azure Databricks
- Provision Azure Databricks workspaces and clusters
- Work with notebooks in Azure Databricks
- Module 2: Work with data in Azure Databricks
After completing this module, you will be able to:
- Understand dataframes
- Query dataframes
- Visualize data
- Module 3: Prepare data for machine learning with Azure Databricks
After completing this module, you will be able to:
- Understand machine learning concepts
- Perform data cleaning
- Perform feature engineering
- Perform data scaling
- Perform data encoding
- Module 4: Train a machine learning model with Azure Databricks
After completing this module, you will be able to:
- Understand Spark ML
- Train and validate a model
- Use other machine learning frameworks
- Module 5: Use MLflow to track experiments in Azure Databricks
After completing this module, you will be able to:
- Understand capabilities of MLflow
- Use MLflow terminology
- Run experiments
- Module 6: Manage machine learning models in Azure Databricks
After completing this module, you will be able to:
- Describe considerations for model management
- Register models
- Manage model versioning
- Module 7: Track Azure Databricks experiments in Azure Machine Learning
After completing this module, you will be able to:
- Describe Azure Machine Learning
- Run Azure Databricks experiments in Azure Machine Learning
- Log metrics in Azure Machine Learning with MLflow
- Run Azure Machine Learning pipelines on Azure Databricks compute
- Module 8: Deploy Azure Databricks models in Azure Machine Learning
After completing this module, you will be able to:
- Describe considerations for model deployment
- Plan for Azure Machine Learning deployment endpoints
- Deploy a model to Azure Machine Learning
- Troubleshoot model deployment
- Module 9: Tune hyperparameters with Azure Databricks
After completing this module, you will be able to:
- Understand hyperparameter tuning and its role in machine learning.
- Learn how to use the two open-source tools - automated MLflow and Hyperopt - to automate the process of model selection and hyperparameter tuning.
- Module 10: Distributed deep learning with Horovod and Azure Databricks
After completing this module, you’ll be able to:
- Understand what Horovod is and how it can help distribute your deep learning models.
- Use HorovodRunner in Azure Databricks for distributed deep learning.