Databricks Machine Learning Associate Question 641: You are tasked with building a machine learning model that performs... | Databricks Machine Learning Associate Dumps

Databricks Machine Learning Associate practice question 641 of 656

Databricks Certified Machine Learning Associate. Associate level, Databricks. Free question with the correct answer and a full explanation.

Databricks Machine Learning Associate Question 641

Select 3

You are tasked with building a machine learning model that performs real-time predictions for streaming data using Delta Live Tables in Databricks. Which of the following steps are required to implement streaming inference in this scenario?

A
Define a Delta Live Table pipeline with a streaming input dataset.
B
Use a machine learning model hosted as a UDF (user-defined function) to apply predictions to the streaming data.
C
Configure the Delta Live Table pipeline to run in a batch mode to process historical data.
D
Persist the streaming prediction outputs into a Delta table for downstream consumption.
E
Set the pipeline mode to 'triggered' for continuous processing of the streaming data.

Show answer and explanation

Correct answers: A, B, D

Explanation

To implement streaming inference with Delta Live Tables, you must set up a pipeline that consumes streaming data, applies predictions in real time using a machine learning model (typically as a UDF), and stores the prediction results in a Delta table for further use. Batch processing and 'triggered' mode are not suitable for real-time streaming scenarios, as they do not support continuous data processing.

A. Correct.
Defining a Delta Live Table pipeline with a streaming input dataset is necessary to process real-time data streams in Delta Live Tables.
B. Correct.
Using a machine learning model hosted as a UDF is required to apply predictions to the incoming streaming data.
C. Incorrect.
Configuring the Delta Live Table pipeline to run in batch mode is not suitable for real-time streaming inference, as batch mode processes data in fixed intervals rather than continuously.
D. Correct.
Persisting the streaming prediction outputs into a Delta table ensures that the results are saved and available for downstream analytics or applications.
E. Incorrect.
Setting the pipeline mode to 'triggered' is not required for streaming inference; instead, 'continuous' mode should be used for real-time processing.

Databricks Machine Learning Associate practice question 641 of 656

Databricks Machine Learning Associate Question 641

Take a Databricks Machine Learning Associate practice test under exam conditions

More Databricks Machine Learning Associate practice questions