You plan to run a script as an experiment using a Script Run Configuration. The script uses modules from the scipy library as well as several Python packages that are not typically installed in a…

Question

Options

• ADo not specify an environment in the run configuration for the experiment. Run the experiment by
• BCreate a virtual machine (VM) with the required Python configuration and attach the VM as a
• CCreate and register an Environment that includes the required packages. Use this Environment
• DCreate a config.yaml file defining the conda packages that are required and save the file in the
• EAlways run the experiment with an Estimator by using the default packages.

• B
  11% (2)
• C
  79% (15)
• D
  5% (1)
• E
  5% (1)

Explanation

Creating and registering an Azure ML Environment that includes all required packages (scipy and other custom libraries) is the best approach for running experiments on both local and remote compute with minimal administrative effort. A registered Environment is reusable, version-controlled, and can be consistently applied to any compute target-local or remote. Azure ML will build and cache the Docker image or conda environment so remote clusters use the same configuration automatically. Option A fails because without specifying an environment, missing packages will cause errors on remote compute. Option B requires unnecessary VM management. Option D (config.yaml) only works for conda packages and has limited portability. Option E (Estimators) is a legacy approach and does not guarantee all packages are available.

No community discussion yet for this question.