| import logging |
| import sys |
| |
| from hamilton import base, driver |
| from hamilton.plugins import polars_implementations |
| |
| logging.basicConfig(stream=sys.stdout) |
| |
| # Create a driver instance. |
| adapter = base.SimplePythonGraphAdapter( |
| result_builder=polars_implementations.PolarsDataFrameResult() |
| ) |
| config = { |
| "base_df_location": "dummy_value", |
| } |
| import my_functions # where our functions are defined |
| |
| dr = driver.Driver(config, my_functions, adapter=adapter) |
| # note -- currently the result builder does not handle mixed outputs, e.g. Series and scalars. |
| output_columns = [ |
| "spend", |
| "signups", |
| "avg_3wk_spend", |
| "spend_per_signup", |
| "spend_zero_mean_unit_variance", |
| ] |
| # let's create the dataframe! |
| df = dr.execute(output_columns) |
| print(df) |
| |
| # To visualize do `pip install "sf-hamilton[visualization]"` if you want these to work |
| # dr.visualize_execution(output_columns, './polars', {"format": "png"}) |
| # dr.display_all_functions('./my_full_dag.dot') |
