Databricks Machine Learning Associate exam dumps

Databricks Machine Learning Associate practice question 645 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 645

Select 3

A data engineering team is designing a real-time machine learning inference pipeline using Delta Live Tables (DLT) in Databricks. They want to process streaming input data, apply a pre-trained model for predictions, and write the results to a Delta table. Which of the following steps are necessary to perform streaming inference using Delta Live Tables?

  1. A

    Define a DLT pipeline with the input data source configured as a streaming source.

  2. B

    Use a Python function with a pre-trained model to apply predictions in a @dlt.view or @dlt.table decorator.

  3. C

    Set the DLT pipeline to 'Triggered' mode to ensure continual processing of the streaming data.

  4. D

    Write the inference results to a Delta table using a @dlt.table decorator.

  5. E

    Ensure the pre-trained model is stored in a Databricks-hosted MLflow Model Registry for access during inference.

Show answer and explanation

Correct answers: A, B, D

Explanation

Delta Live Tables support real-time streaming inference by allowing users to define streaming input sources, process the data using Python functions with pre-trained models, and write the results to Delta tables using decorators like @dlt.table. While additional tools like the MLflow Model Registry can enhance the process, they are not strictly necessary for streaming inference in DLT.

  • A. Correct.

    This is correct. Streaming inference requires a streaming source for real-time data ingestion, which DLT supports by configuring the input data source as a streaming source.

  • B. Correct.

    This is correct. Applying a pre-trained model for predictions typically involves using a Python function within a @dlt.view or @dlt.table decorator in the DLT pipeline.

  • C. Incorrect.

    This is incorrect. 'Triggered' mode is not required for streaming inference. DLT supports continuous processing for streaming workloads without needing to set this mode explicitly.

  • D. Correct.

    This is correct. Writing the inference results to a Delta table is a key step in streaming inference, and DLT facilitates this using a @dlt.table decorator.

  • E. Incorrect.

    This is incorrect. While storing the model in the MLflow Model Registry could be helpful for model management, it is not a mandatory requirement for performing streaming inference in DLT.

Timed practice exam

Take a Databricks Machine Learning Associate practice test under exam conditions

48 questions in 90 minutes, drawn from this bank, with a score report and a per-question review when you finish.

Start timed exam

More Databricks Machine Learning Associate practice questions

  1. 646.You have trained a machine learning model in Databricks and registered it in the MLflow Model...
  2. 647.A data science team has trained a machine learning model in Databricks and wants to deploy it for...
  3. 648.You are tasked with deploying a machine learning model for real-time inference using Databricks....
  4. 649.You have trained a machine learning model for fraud detection using Databricks MLflow. Your team...
  5. 650.You are working on deploying a machine learning model for real-time inference using Databricks....
Back to the Databricks Machine Learning Associate exam dumps