Exam DP-100 topic 2 question 62 discussion

given answer is correct

Correct, the answer is: Dataset and Tabular.from_delimited_files To make CSV files available for use in Azure ML experiments and data processing pipelines, and specifically for them to be loaded directly into pandas dataframes, the appropriate method from the Dataset class is Tabular.from_delimited_files. - Tabular.from_delimited_files is designed for structured files like CSVs and will load the data into a format that can be easily converted to pandas dataframes, which is exactly the requirement. Wrong answers: - File.from_files is used when you are working with file datasets that are which a collection of references to the files directly without loading their contents into a structured format, rather than providing a dataset that represents the data in a tabular format like CSV. - Tabular.from_pandas_dataframe is used to create a tabular dataset from an existing pandas dataframe in memory, not from files in a datastore.

Sorry for the previous response. I've double-checked and updated the code. The answer is correct. from azureml.core import Workspace, Dataset ws = Workspace.from_config() blob_ds = ws.get_default_datastore () target_data = [(blob ds, 'data/files/archive/*.csv')] data1 = Dataset.Tabular.from_delimited_files(path=target_data) registered_datal datal.register(workspaws, name data1')

The mentioned answer is partially correct. The actual code is given below. from azureml.core import Workspace, Dataset ws = Workspace.from_config() blob_ds = ws.get_default_datastore () target_data = [(blob ds, 'data/files/archive/*.csv')] data1 = Dataset.Files.from_files(path=target_data) registered_datal datal.register(workspaws, name datal')

It seems correct https://github.com/MicrosoftDocs/azure-docs/blob/main/articles/machine-learning/v1/how-to-create-register-datasets.md#create-a-tabulardataset

correct