Answer: image

The correct answer is B (image). According to the Azure Machine Learning Python SDK v2 documentation and the community discussion, when defining an environment from a Docker image, you should use the 'image' parameter to specify the Docker image name. The 'build' parameter is used when you want to define an environment from a Dockerfile, not from an existing Docker image. The community discussion shows that option B has strong support with references to official Microsoft documentation, while option C (build) is incorrect for this specific scenario of using an existing Docker image. The other options (properties and conda_file) are not relevant for defining an environment directly from a Docker image.

Author: LeetQuiz Editorial Team