DURATION: 4 Day.
Time Division: – Break: 15 + 45 + 15 minutes
Course Outcomes:
- Design and prepare a machine learning solution.
- Explore data and train models.
- Prepare a model for deployment.
- Deploy and retrain a model.
Important Note:
- Courseware – Reference material/ppt along with lab files/exercises will be provided.
- Azurepass/Virtual Machine will be provided only during the training time to perform the labs.
Module 1: Explore and configure the Azure Machine Learning workspace
- Explore Azure Machine Learning workspace resources and assets.
- Lab: Explore Azure Machine Learning workspace resources and assets.
- Explore developer tools for workspace interaction.
- Lab: Explore developer tools for workspace interaction.
- Work with compute targets in Azure Machine Learning.
- Lab: Work with compute targets in Azure Machine Learning.
- Work with environments in Azure Machine Learning.
- Lab: Work with environments in Azure Machine Learning.
Module 2: Work with Data in Azure Machine Learning
- Make data available in Azure Machine Learning.
- Lab: Make data available in Azure Machine Learning.
Module 3: Experiment with Azure Machine Learning
- Find the best classification model with Automated Machine Learning.
- Lab: Train a model with the Azure Machine Learning Designer.
- Lab: Find the best classification model with Automated Machine Learning.
- Track model training in Jupiter notebooks with MLflow.
- Lab: Track model training in notebooks with MLflow.
Module 4: Train models with scripts in Azure Machine Learning
- Run a training script as a command job in Azure Machine Learning.
- Lab: Use MLflow to track training jobs.
- Track model training with MLflow in jobs.
- Lab: Track model training with MLflow in jobs.
Module 5: Optimize model training with Azure Machine Learning
- Run pipelines in Azure Machine Learning.
- Lab: Run pipelines in Azure Machine Learning.
- Perform hyper parameter tuning with Azure Machine Learning.
- Lab: Perform hyper parameter tuning with Azure Machine Learning.
Module 6: Deploy and consume models with Azure Machine Learning
- Deploy a model to a managed online endpoint.
- Lab: Log and register models with MLflow.
- Lab: Compare and evaluate models.
- Lab: Deploy a model to a managed online endpoint.
- Deploy a model to a batch endpoint.
- Lab: Deploy a model to a batch endpoint.