Custom Model Serving MACHINE-LEARNING-PROFESSIONAL Practice Questions (Page 4)

Questions 16–17

1. 16application · medium

   A company has deployed a custom PyFunc model to Databricks Model Serving. They need to integrate it with a web application that sends HTTP requests. Which interface should the web application use to get predictions?

   • AThe Databricks Filesystem API.
   • BThe Databricks REST API endpoint for the serving endpoint.
   • CThe Databricks SQL API.
   • DThe MLflow model registry API.
   Reveal Solution Discussion Select an answer first
2. 17expert · hard

   A team has a custom PyFunc model that uses a library that is not available in the default Databricks runtime. They have logged the model with a conda environment that includes the library. However, when they deploy the model to Model Serving, it fails with an import error. What is the most likely cause?

   • AThe library is too large to be installed in the serving environment.
   • BThe model was logged with a different Python version than the serving environment.
   • CThe library is not compatible with the serving infrastructure.
   • DThe conda environment was not properly captured during model logging.
   Reveal Solution Discussion Select an answer first