AI-102 Question 306
Select 3You are designing a solution to process data for a machine learning model in Azure. The data is stored in multiple sources, including an Azure Blob Storage container, an Azure SQL Database, and a local file on your on-premises network. You need to import the data into Azure Machine Learning for preprocessing and model training. Which of the following sources can you directly import from using Azure Machine Learning's data import capabilities?
- A
Azure Blob Storage
- B
Azure SQL Database
- C
Local file on on-premises network
- D
Azure Table Storage
- E
GitHub repository
Show answer and explanation
Correct answers: A, B, D
Explanation
Azure Machine Learning supports importing data from several Azure-based sources, such as Azure Blob Storage, Azure SQL Database, and Azure Table Storage. Local files or repositories like GitHub must first be uploaded to a supported Azure service before they can be imported into Azure Machine Learning.
- A. Correct.
Azure Blob Storage is a supported data source for Azure Machine Learning. You can directly import data from it using the AzureML SDK or the Azure portal.
- B. Correct.
Azure SQL Database is supported as a data source for Azure Machine Learning. You can use an AzureML Dataset to connect and query data from the database.
- C. Incorrect.
Local files on an on-premises network are not directly supported for import into Azure Machine Learning. You would first need to upload the file to an Azure storage service.
- D. Correct.
Azure Table Storage is a supported data source that can be accessed and imported into Azure Machine Learning.
- E. Incorrect.
GitHub repositories are not directly supported as data import sources for Azure Machine Learning. You would need to clone the repository and upload the data to Azure storage or another supported data source.