load_join.py

import pyspark.sql as ps
from hamilton.function_modifiers import load_from, value, source

@load_from.csv(path=value("data_1.csv"), inject_="raw_data_1", spark=source("spark"))
@load_from.parquet(path=value("data_2.parquet"), inject_="raw_data_2", spark=source("spark"))
def all_initial_data(raw_data_1: ps.DataFrame, raw_data_2: ps.DataFrame) -> ps.DataFrame:
    """Combines the two loaded dataframes"""
    return _custom_join(raw_data_1, raw_data_2)