Designing and Implementing a Data Science Solution on Azure — Question 71

You are developing a machine learning model by using Azure Machine Learning. You are using multiple text files in tabular format for model data.

You have the following requirements:

• You must use AutoMLjobs to train the model.
• You must use data from specified columns.
• The data concept must support lazy evaluation.

You need to load data into a Pandas dataframe.

Which data concept should you use?

Answer options

• A. Data asset
• B. URI
• C. Datastore
• D. MLTable

Correct answer: D

Explanation

The correct answer is MLTable because it allows for lazy evaluation and can be configured to load specific columns necessary for AutoML jobs. The other options like Data asset and Datastore do not inherently support lazy evaluation in the same way, and URI is simply a reference rather than a data loading concept.