SKILL BENCHMARK
Azure AI Fundamentals: Machine Learning on Microsoft Azure Competency (Intermediate Level)
- 23m
- 23 questions
The Machine Learning on Microsoft Azure Competency benchmark measures your ability to recall and identify common machine learning types, core machine learning concepts such as identifying features and labels in a dataset, core tasks in creating machine learning solutions, and capabilities of no-code machine learning with Azure Learning Studio. A learner who scores high on this benchmark demonstrates that they have the necessary knowledge and skills to build machine learning solutions on the Azure platform.
Topics covered
- add and use a regression training model in ML Designer
- add a Scoring model component in the ML Designer
- add normalizing and cleaning components for data in ML Designer
- configure and use a classification model in ML Designer
- create and configure an Azure Machine Learning workspace
- create and use a dataset component in Azure ML Designer
- deploy an automated ML model as a predictive service
- deploy the model as a predictive service
- describe a dataset and how they are created and managed
- describe model evaluation types like MAE and R2
- describe the limitations and features of automated ML model training
- describe the machine learning services provided by Azure
- identify the process and functions of Azure ML Studio for creating, running, and maintaining AI workloads
- ingest data from an Azure Storage source
- inspect the Azure ML Studio sidebar components used for creating machine learning workflows
- interpret the results and logs form running a Clustering model
- interpret the results from running a Classification model
- investigate the logs and results that are significant when running a Regression model
- manage pipelines in the Azure ML Studio interface
- manipulate and configure datasets and experiments, including for other team members, in Azure ML Studio
- test the predictive service from an external app
- use a clustering model in ML Designer
- use the automated ML model to create an experiment that will automatically find the best-fit model
