DP-3007: Train and deploy a machine learning model with Azure Machine Learning

Course DP-3007: Train and deploy a machine learning model with Azure Machine Learning

To obtain this Microsoft Applied Skills credential, students demonstrate the ability to train and manage machine learning models with Azure Machine Learning.

Candidates for this credential should be familiar with Azure services and have experience with Azure Machine Learning and MLflow . Candidates should also have experience performing machine learning-related tasks using Python .

Intermediate - Azure, Azure Machine Learning - AI Engineer, Data Engineer, Developer, Data Scientist - Machine Learning

Objectives of the DP-3007 training

• Setting up a development environment in Azure Machine Learning
• Prepare data for model training
• Create and configure a model training script as a command job
• Manage artifacts using MLflow
• Implement a model for real-time consumption

Course content DP-3007

Module 1 Making data available in Azure Machine Learning

• Description of the URIs
• Creating a data warehouse
• Create a data resource
• Exercise: Make data available in Azure Machine Learning

Module 2 Working with Process Targets in Azure Machine Learning

• Choosing the appropriate process destination
• Creating and using a process instance
• Creating and using a process cluster
• Exercise: Working with process resources

Module 3 Working with environments in Azure Machine Learning

• Information about the environments
• Exploration and use of maintained environments
• Creation and use of custom environments
• Exercise: Working with environments

Module 4 Running a training script as a command job in Azure Machine Learning

• Converting a notebook into a script
• Running a script as a command job
• Using parameters in a command job
• Exercise: Running a training script as a command job

Module 5 Monitoring model training with MLflow in jobs

• Tracking metrics with MLflow
• Visualization of metrics and evaluation of models
• Exercise: Using MLflow to track training jobs

Module 6: Registering an MLFlow Model in Azure Machine Learning

• Registering models with MLflow
• Description of the MLflow model format
• Registering an MLflow model
• Exercise: registering models with MLflow

Module 7 Implementation of a model at a managed online connection point

• Exploration of managed online connection points
• Implementing an MLflow model on a managed online endpoint
• Implementation of a model at a managed online connection point
• Testing managed online connection points
• Exercise: Implementing an MLflow model at an online connection point

Prerequisites

Familiarity with Azure services and experience with Azure Machine Learning and MLflow are recommended. Additionally, experience performing machine learning tasks using Python is essential.

Language

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

Microsoft Applied Skills

This course is part of the Microsoft Applied Skills Credentials.

To obtain this Microsoft Applied Skills credential, students demonstrate the ability to train and manage machine learning models with Azure Machine Learning.

Applied Skills: Explore all credentials in one guide