Databricks Machine Learning Associate Question 621
Select 3You are tasked with deploying a trained machine learning model on Databricks to serve predictions in real time. Which of the following steps are necessary to successfully deploy and serve the model using MLflow?
- A
Register the model in the MLflow Model Registry.
- B
Set up a Databricks cluster with GPU support for model serving.
- C
Enable the Databricks Model Serving feature and select the registered model.
- D
Package the model as a Docker container before deployment.
- E
Test the deployed endpoint using sample input data.
Show answer and explanation
Correct answers: A, C, E
Explanation
To deploy a model using Databricks Model Serving, you must register the model in the MLflow Model Registry, enable the Model Serving feature, and test the endpoint to validate its performance. GPU clusters and manual Docker packaging are not required when using Databricks' managed serving capabilities.
- A. Correct.
Registering the model in the MLflow Model Registry is required to manage the model lifecycle and make it available for deployment.
- B. Incorrect.
While GPU clusters might be necessary for certain types of models (e.g., deep learning), it is not a mandatory step for deploying a model with Databricks and MLflow.
- C. Correct.
Enabling the Model Serving feature and selecting the registered model is essential to deploying the model for real-time serving in Databricks.
- D. Incorrect.
It is not necessary to package the model as a Docker container when using Databricks Model Serving, as this service abstracts deployment complexities.
- E. Correct.
Testing the deployed endpoint with sample input data is a best practice to ensure the model is serving predictions as expected.