Exam DP-100 topic 5 question 30 discussion

It seems the correct answers should be OutputFileDatasetConfig and as_input(). It is clearly explained on Microsoft Learn: https://docs.microsoft.com/en-us/training/modules/create-pipelines-in-aml/3-pipeline-data

Class: OutputDatasetConfig Method: as_named_input

OutputFileDatasetConfig as_named_input https://learn.microsoft.com/en-us/azure/machine-learning/how-to-move-data-in-out-of-pipelines?view=azureml-api-1 from azureml.data import OutputFileDatasetConfig dataprep_output = OutputFileDatasetConfig() input_dataset = Dataset.get_by_name(workspace, 'raw_data') dataprep_step = PythonScriptStep( name="prep_data", script_name="dataprep.py", compute_target=cluster, arguments=[input_dataset.as_named_input('raw_data').as_mount(), dataprep_output] )

Exactly this question was on exam 07/March/2023

Class: OutputDatasetConfig Method: as_named_input You could use the as_input method instead of as_named_input if you do not need to pass the output from the first step as a named input in the second step. The as_input method is used to pass an output from a previous step as an input to a subsequent step. However, when you use as_input, the input will not have a name and it will be treated as a regular positional argument to the script. In contrast, when you use as_named_input, the input will have a name and can be consumed using that name as an argument to the script. In the case of passing temporary data from the first step to the second step, it is generally recommended to use as_named_input so that the input can be easily identified and consumed in the second step. Therefore, you could use as_input instead of as_named_input, but as_named_input is the more appropriate method to use in this case.

Microsoft mixed up their own stuff here. It is the PipelineData object used for temporary data storage produced by one step of a pipeline to another step, creating a data dependency! See: https://learn.microsoft.com/en-us/python/api/azureml-pipeline-core/azureml.pipeline.core.pipelinedata?view=azure-ml-py Even in the mentioned " https://learn.microsoft.com/en-us/training/modules/create-pipelines-in-aml/3-pipeline-data" they commented "PipelineData" but wrote "OutputFileDatasetConfig", take a look at the code.