azureml_dataprep_csv_load_filter_transform_to_pandas.py

1import azureml.dataprep as dprep
2
3# Load data from a delimited file (e.g., a CSV)
4# For this example, we assume a local CSV file exists; 
5# you can replace the path with a URL or datastore path.
6data_flow = dprep.read_csv(path='./data.csv')
7
8# Show the first 5 rows of the data
9print("Initial data preview:")
10print(data_flow.head(5))
11
12# Perform a transformation: Filter rows where a specific column's value is greater than 0
13# And keep only specific columns
14data_flow = data_flow.filter(dprep.col('Column1') > 0)
15data_flow = data_flow.keep_columns(['Column1', 'Column2'])
16
17# Map values or change types (e.g., change Column1 to Decimal)
18data_flow = data_flow.to_number('Column1')
19
20# Convert the DataFlow into a Pandas DataFrame for further analysis
21df = data_flow.to_pandas_dataframe()
22
23print("\nTransformed data preview:")
24print(df.head())