Databricks Machine Learning Associate Question 444: You are tasked with scoring a machine learning model using features... | Databricks Machine Learning Associate Dumps

Databricks Machine Learning Associate practice question 444 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 444

Select 3

You are tasked with scoring a machine learning model using features stored in a feature store table in Databricks. The feature table contains the necessary features for scoring, and the model has been registered in the MLflow Model Registry. Which of the following steps should you perform to score the model using data from the feature store?

A
Use the FeatureStoreClient to load the feature data into a DataFrame.
B
Retrieve the model from the MLflow Model Registry using the mlflow.pyfunc.load_model function.
C
Join the feature store table with the input data using the fs.read_table method.
D
Use the FeatureStoreClient.score_batch method to score the model with the features.
E
Directly pass the feature store table path to the mlflow.pyfunc.load_model function for scoring.

Show answer and explanation

Correct answers: A, B, D

Explanation

To score a machine learning model using features from a feature store table in Databricks, you need to first load the feature data using the FeatureStoreClient and access the model from the MLflow Model Registry. Once the features and model are available, you can use the FeatureStoreClient.score_batch method to score the model in a batch setting. Additional steps, such as joining feature tables, are only required if the feature data needs preprocessing or enrichment.

A. Correct.
Correct: The FeatureStoreClient is used to access and load feature data from the feature store into a DataFrame, which can then be used for model scoring.
B. Correct.
Correct: The model must be retrieved from the MLflow Model Registry using mlflow.pyfunc.load_model or a similar method to load the model for scoring.
C. Incorrect.
Incorrect: While you can read feature data using fs.read_table, joining the table is not a necessary step for scoring if the feature data is already prepared.
D. Correct.
Correct: The FeatureStoreClient.score_batch method provides a streamlined way to score a model using features from a feature store in a batch setting.
E. Incorrect.
Incorrect: The feature store table path cannot be directly passed to mlflow.pyfunc.load_model for scoring. The model and features must be handled separately.

Databricks Machine Learning Associate practice question 444 of 656

Databricks Machine Learning Associate Question 444

Take a Databricks Machine Learning Associate practice test under exam conditions

More Databricks Machine Learning Associate practice questions