Answer-first summary for fast verification
Answer: model loader
The question describes a scenario where a custom model cannot be serialized in Pickle format, has complex source code, and requires packaging the Python library with the model. According to Microsoft documentation and the community discussion (with 100% consensus and upvoted references), the model loader approach is specifically designed for such cases. The model loader method allows packaging all required artifacts, custom loading logic, and complex source code as a single unit, making it the optimal choice when Pickle serialization is not suitable and the model requires custom handling. Other options like artifacts, model wrapper, or custom signatures don't provide the comprehensive packaging and loading capabilities needed for this complex custom model scenario.
Author: LeetQuiz Editorial Team
Ultimate access to all questions.
You are managing an Azure Machine Learning workspace and have built a custom model that must be logged with MLflow. The model has the following characteristics:
You need to create a custom model flavor to enable logging with MLflow.
What should you use?
A
model loader
B
artifacts
C
model wrapper
D
custom signatures
No comments yet.