Designing and implementing a data science solution on Azure

Course Overview:

Learn how to operate machine learning solutions at cloud scale using Azure Machine Learning. This course teaches you to leverage your existing knowledge of Python and machine learning to manage data ingestion and preparation, model training and deployment, and machine learning solution monitoring with Azure Machine Learning and MLflow.

Course Objectives:

• Learn how to design a data ingestion solution for training data used in machine learning projects.
• Learn how to design a model training solution for machine learning projects.
• Learn how to design a model deployment solution and how the requirements of the deployed model can affect the way you train a model.
• Learn about machine learning operations or MLOps to bring a model from development to production. Identify options for monitoring and retraining when preparing a model for production.
• Learn what Azure Machine Learning is, and get familiar with all its resources and assets.
• Learn how you can interact with the Azure Machine Learning workspace. You can use the Azure Machine Learning studio, the Python SDK (v2), or the Azure CLI (v2).
• Learn about how to connect to data from the Azure Machine Learning workspace. You’re introduced to datastores and data assets.
• Learn how to work with compute targets in Azure Machine Learning. Compute targets allow you to run your machine learning workloads. Explore how and when you can use a compute instance or compute cluster.
• Learn how to use environments in Azure Machine Learning to run scripts on any compute target.
• Learn how to find the best classification model with automated machine learning (AutoML). You’ll use the Python SDK (v2) to configure and run an AutoML job.
• Learn how to use MLflow for model tracking when experimenting in notebooks.
• Learn how to convert your code to a script and run it as a command job in Azure Machine Learning.
• Learn how to track model training with MLflow in jobs when running scripts.
• Learn how to perform hyperparameter tuning with a sweep job in Azure Machine Learning.
• Learn how to create and use components to build pipeline in Azure Machine Learning. Run and schedule Azure Machine Learning pipelines to automate machine learning workflows.
• Learn how to log and register an MLflow model in Azure Machine Learning.
• Explore model explanations, error analysis, counterfactuals, and causal analysis by creating a Responsible AI dashboard. You’ll create and run the pipeline in Azure Machine Learning using the Python SDK v2 to generate the dashboard.
• Learn how to deploy models to a managed online endpoint for real-time inferencing.
• Learn how to deploy models to a batch endpoint. When you invoke a batch endpoint, you’ll trigger a batch scoring job.

Who Should Attend?

This course is designed for data scientists with existing knowledge of Python and machine learning frameworks like Scikit-Learn, PyTorch, and Tensorflow, who want to build and operate machine learning solutions in the cloud.

Course Prerequisites

There are no prerequisites for this course.

Course Content:

Module 1: Design a data ingestion strategy for machine learning projects
Introduction
Identify your data source and format
Choose how to serve data to machine learning workflows
Design a data ingestion solution
Exercise: Design a data ingestion strategy
Knowledge check
Summary

Module 2: Design a machine learning model training solution
Introduction
Identify machine learning tasks
Choose a service to train a machine learning model
Decide between compute options
Exercise: Design a model training strategy
Knowledge check
Summary

Module 3: Design a model deployment solution
Introduction
Understand how model will be consumed
Decide on real-time or batch deployment
Exercise – Design a deployment solution
Summary

Module 4: Design a machine learning operations solution
Introduction
Explore an MLOps architecture
Design for monitoring
Design for retraining
Knowledge check
Summary

Module 5: Explore Azure Machine Learning workspace resources and assets
Introduction
Create an Azure Machine Learning workspace
Identify Azure Machine Learning resources
Identify Azure Machine Learning assets
Train models in the workspace
Exercise – Explore the workspace
Knowledge check
Summary

Module 6: Explore developer tools for workspace interaction
Introduction
Explore the studio
Explore the Python SDK
Explore the CLI
Exercise – Explore the developer tools
Knowledge check
Summary

Module 7: Make data available in Azure Machine Learning
Introduction
Understand URIs
Create a datastore
Create a data asset
Exercise – Make data available in Azure Machine Learning
Knowledge check
Summary

Module 8: Work with compute targets in Azure Machine Learning
Introduction
Choose the appropriate compute target
Create and use a compute instance
Create and use a compute cluster
Exercise – Work with compute resources
Knowledge check
Summary

Module 9: Work with environments in Azure Machine Learning
Introduction
Understand environments
Explore and use curated environments
Create and use custom environments
Exercise – Work with environments
Knowledge check
Summary

Module 10: Find the best classification model with Automated Machine Learning
Introduction
Preprocess data and configure featurization
Run an Automated Machine Learning experiment
Evaluate and compare models
Exercise – Find the best classification model with Automated Machine Learning
Knowledge check
Summary

Module 11: Track model training in Jupyter notebooks with MLflow
Introduction
Configure MLflow for model tracking in notebooks
Train and track models in notebooks
Exercise – Track model training
Knowledge check
Summary

Module 12: Run a training script as a command job in Azure Machine Learning
Introduction
Convert a notebook to a script
Run a script as a command job
Use parameters in a command job
Exercise – Run a training script as a command job
Knowledge check
Summary

Module 13: Track model training with MLflow in jobs
Introduction
Track metrics with MLflow
View metrics and evaluate models
Exercise – Use MLflow to track training jobs
Knowledge check
Summary

Module 14: Perform hyperparameter tuning with Azure Machine Learning
Introduction
Define a search space
Configure a sampling method
Configure early termination
Use a sweep job for hyperparameter tuning
Exercise – Run a sweep job
Knowledge check
Summary

Module 15: Run pipelines in Azure Machine Learning
Introduction
Create components
Create a pipeline
Run a pipeline job
Exercise – Run a pipeline job
Knowledge check
Summary

Module 16: Register an MLflow model in Azure Machine Learning
Introduction
Log models with MLflow
Understand the MLflow model format
Register an MLflow model
Exercise – Log and register models with MLflow
Knowledge check
Summary

Module 17: Create and explore the Responsible AI dashboard for a model in Azure Machine Learning
Introduction
Understand Responsible AI
Create the Responsible AI dashboard
Evaluate the Responsible AI dashboard
Exercise – Explore the Responsible AI dashboard
Knowledge check
Summary

Module 18: Deploy a model to a managed online endpoint
Introduction
Explore managed online endpoints
Deploy your MLflow model to a managed online endpoint
Deploy a model to a managed online endpoint
Test managed online endpoints
Exercise – Deploy an MLflow model to an online endpoint
Knowledge check
Summary

Module 19: Deploy a model to a batch endpoint
Introduction
Understand and create batch endpoints
Deploy your MLflow model to a batch endpoint
Deploy a custom model to a batch endpoint
Invoke and troubleshoot batch endpoints
Exercise – Deploy an MLflow model to a batch endpoint
Knowledge check
Summary