SKILL BENCHMARK
DP-100: Design and Prepare Azure Machine Learning Solutions Competency (Intermediate Level)
- 37m
- 37 questions
The Design and Prepare Azure Machine Learning Solutions Competency (Intermediate Level) benchmark measures your ability to design a machine learning (ML) solution and manage an Azure Machine Learning workspace and its data. You will be evaluated on your skills in managing and registering data stores, recognizing various Azure Data Platform services, and creating and managing compute resources. A learner who scores high on this benchmark demonstrates that they have good experience in managing Azure Machine Learning workspaces and data.
Topics covered
- configure metrics for a storage account to monitor throughput
- create an Azure container and an Azure file share
- create an Azure Cosmos DB using the Azure portal
- create an Azure Data Lake Storage Gen2 solution
- create an Azure Machine Learning workspace and resource group using Azure Portal and use Azure Machine Learning Studio to create a compute resource, and clone a notebook
- create and manage a compute cluster in the Azure Machine Learning Workspace
- create and manage a compute instance in the Azure Machine Learning workspace
- create and register data stores in Machine Learning Studio for Azure blob container, Azure file share, and Azure Data Lake Storage Gen 2
- create a script to train a model, add parameters to the script, and run the object to get the training model
- create Azure Table storage and Queue storage
- create Python scripts to run an experiment, log metrics, and retrieve and view logged metrics
- describe Azure Cosmos DB and the API models that can be used with it
- describe how Azure Stream Analytics is used to process streaming data
- describe how data can be stored using Azure Data Storage
- describe how to create an Azure storage account and how many you will need
- describe Jupyter Notebooks and how they are used by data scientists to perform data analysis
- describe strategies and mechanisms for securing storage account data
- describe the Azure SQL Database and when to use it
- describe the Azure Synapse Analytics platform and how it is used for data warehousing and big data analytics
- describe the concepts and features of dynamic data masking using the Azure portal
- describe the concepts and features of the encrypting data at rest and in motion in Azure
- describe the differences between structured and unstructured data types and how they can be stored in Azure
- describe the features and best practices for working with Azure data blobs
- describe the features and components of the Azure Machine Learning workspace
- describe the features of Azure Databricks
- describe the features of Azure HDInsight for ingesting, processing, and analyzing big data
- describe the features of the Azure Data Factory
- describe the features of the Azure Data Lake Storage Gen2 and when to use this storage type
- describe the tools available to create Azure storage accounts
- describe the types of datasets that can be created and then create, register, and use datasets
- describe types of Machine Learning Studio compute targets such as local compute, compute clusters, and attached compute
- differentiate between structured and unstructured data types and recognize how they can be stored in Azure
- run a notebook using Jupyter to work with data, data stores, and datasets
- upload and download blob data to and from an Azure Storage Account
- use a Jupyter Notebook to create a regression model
- use a Jupyter Notebook to perform deep learning using TensorFlow
- use the Azure Machine Learning SDK to run code experiments that log metrics and generate outputs
