Develop a comprehensive understanding of machine learning models, data privacy safeguards, and effective end-to-end management of the machine learning lifecycle at scale using Azure Machine Learning's MLOps capabilities.
-
Understanding the rationale for cloud-based machine learning.
-
Efficiently utilizing workspaces and AzureML Studio.
-
Integrating third-party and open datasets into machine learning pipelines.
-
Managing pipelines and leveraging hyperparameters for improved prediction accuracy.
-
Programmatically creating and managing pipelines using the Azure ML SDK.
-
Automating machine learning processes with Hyperparameter Tuning and AutoML.
-
Authorizing operations for machine learning.
-
Deploying machine learning models in Azure.
-
Consuming and load-testing deployed services and endpoints.
-
Creating batch inference pipelines and publishing them.
-
Applying DevOps concepts for model deployment.
-
Configuring and deploying a cloud-based machine learning production model using Azure.
- Combining all skills acquired in this program for a self-choosen ML project → Capstone project: Heart Failure Prediction with AzureML
- Azure Machine Learning: Azure ML platform, Azure ML pipelines, Model interpretation, Azure ML SDK, Hyperparameter tuning.
- Machine Learning Operations: Model deployment with Azure, Kubernetes security, Deployment testing, Docker, Model evaluation.