You use the Azure Machine Learning service to create a tabular dataset named training_data. You plan to use this dataset in a training script. You create a variable that references the dataset using…

Question

Options

• Aenvironment_definition = {"training_data":training_ds}
• Binputs = [training_ds.as_named_input('training_ds')]
• Cscript_params = {"--training_ds":training_ds}
• Dsource_directory = training_ds

• A
  2% (1)
• B
  85% (35)
• C
  7% (3)
• D
  5% (2)

Explanation

inputs = [training_ds.as_named_input('training_ds')] is the correct property. The inputs parameter accepts a list of named dataset inputs; inside the script, the dataset is accessed via run.input_datasets['training_ds']. Option A (environment_definition) configures the Python environment, not data inputs. Option C (script_params) is used for command-line argument values, but passing a Dataset object directly as a param value is not the correct pattern for tabular dataset access. Option D (source_directory) specifies the folder containing the script, not data.

No community discussion yet for this question.