SKILL BENCHMARK
DP-100: Explore Data and Train Models in Azure Competency (Intermediate Level)
- 24m
- 24 questions
The Explore Data and Train Models in Azure Competency (Intermediate Level) benchmark measures your ability to explore data by using data assets and stores and create and train models. You will be evaluated on your skills in using automated machine learning (ML) with Azure Machine Learning. A learner who scores high on this benchmark demonstrates that they have good experience in exploring and preparing datasets for machine learning and automating machine learning in Azure for ML solution development.
Topics covered
- create a K-means clustering model in Azure Machine Learning Studio
- create and use a compute resource
- deploy a classification model based inference pipeline that can be used by clients
- deploy an clustering model based inference pipeline that can be used by clients
- deploy a regression model-based inference pipeline that can be used by clients
- describe datasets and how to manipulate data for those datasets
- describe how features are selected and used from datasets in AI algorithms
- describe how supervised machine learning models use labeled data, are simpler to build, and have more accurate results
- describe how unsupervised machine learning models discover patterns from unlabeled data and can perform complex processing tasks
- describe metrics for determining the best classification model to use
- describe the best metrics for determining which regression model to use
- evaluate a classification model by using an evaluate model in Azure Machine Learning Studio
- evaluate a clustering model by using an evaluate model in Azure Machine Learning Studio
- evaluate a regression model by using an evaluate model in Azure Machine Learning Studio
- ingest data from an Azure Blob storage resource
- register and signup for an Azure Machine Learning Studio account and access the studio dashboard
- use an existing pipeline to create a new inference pipeline to create a predictive service for a classification model
- use an existing pipeline to create a new inference pipeline to create a predictive service for a clustering model
- use an existing pipeline to create a new inference pipeline to create a predictive service for a regression model
- use a subset of data to train the regression model and run the training pipeline
- use a subset of the data to train the classification model and run the training pipeline
- use the Azure Machine Learning designer to train a classification model
- use the Azure Machine Learning designer to train a clustering model
- use the Azure Machine Learning designer to train a regression model
