nerdexam
Microsoft

DP-100 · Question #143

You are a lead data scientist for a project that tracks the health and migration of birds. You create a multi- class image classification deep learning model that uses a set of labeled bird photograph

The correct answer is D. Register the Azure blob storage containing the bird photographs as a datastore in Azure Machine. Registering the existing Azure blob storage as a datastore in Azure ML creates a logical reference (connection) to the blob container-no data is moved. The training code then accesses images directly from the original location. Option A (Azure Data Lake) and Option E (copy to ML

Design and prepare a machine learning solution

Question

You are a lead data scientist for a project that tracks the health and migration of birds. You create a multi- class image classification deep learning model that uses a set of labeled bird photographs collected by experts. You have 100,000 photographs of birds. All photographs use the JPG format and are stored in an Azure blob container in an Azure subscription. You need to access the bird photograph files in the Azure blob container from the Azure Machine Learning service workspace that will be used for deep learning model training. You must minimize data movement. What should you do?

Options

  • ACreate an Azure Data Lake store and move the bird photographs to the store.
  • BCreate an Azure Cosmos DB database and attach the Azure Blob containing bird photographs
  • CCreate and register a dataset by using TabularDataset class that references the Azure blob
  • DRegister the Azure blob storage containing the bird photographs as a datastore in Azure Machine
  • ECopy the bird photographs to the blob datastore that was created with your Azure Machine

How the community answered

(39 responses)
  • A
    5% (2)
  • C
    3% (1)
  • D
    79% (31)
  • E
    13% (5)

Explanation

Registering the existing Azure blob storage as a datastore in Azure ML creates a logical reference (connection) to the blob container-no data is moved. The training code then accesses images directly from the original location. Option A (Azure Data Lake) and Option E (copy to ML blob datastore) both require moving 100,000 photos. Option B (Cosmos DB) is a NoSQL database and inappropriate for storing image files. Option C uses TabularDataset, which is designed for structured/tabular data, not image files; images require a FileDataset. Registering the datastore (D) satisfies both requirements: access from the workspace and zero data movement.

Topics

#Azure Machine Learning Datastores#Azure Blob Storage#Data Access#Deep Learning Data Preparation

Community Discussion

No community discussion yet for this question.

Full DP-100 Practice