azureml_dataprep_csv_load_filter_transform_to_pandas.py

1import azureml.dataprep as dprep
2
3# Load data from a delimited file (CSV)
4# You can replace the path with a local path or a supported cloud URL
5dataflow = dprep.read_csv(path='https://dprepdata.blob.core.windows.net/demo/CrimeData.csv')
6
7# Perform transformations: 
8# 1. Filter rows to keep only those where 'Primary Type' is 'THEFT'
9# 2. Keep only specific columns
10# 3. Take the top 100 rows
11dataflow = dataflow.filter(dprep.col('Primary Type') == 'THEFT')
12dataflow = dataflow.keep_columns(['Case Number', 'Date', 'Block', 'Primary Type'])
13dataflow = dataflow.head(100)
14
15# Execute the dataflow and pull the results into a Pandas DataFrame
16df = dataflow.to_pandas_dataframe()
17
18# Display the first few rows of the resulting DataFrame
19print(df.head())