benchmark_sample_data.py

import random
import pandas as pd
import numpy as np

NUM_DAYS = 3000  # scaled up to show difference

def original():
    df = pd.DataFrame({
        "Date": pd.date_range(start="2024-01-01", periods=NUM_DAYS),
        "Sales": [random.randint(100, 500) for _ in range(NUM_DAYS)],
        "Category": [random.choice(["Electronics", "Clothing", "Home"]) for _ in range(NUM_DAYS)],
        "Profit": [random.randint(10, 100) for _ in range(NUM_DAYS)]
    })
    return df

def optimized():
    df = pd.DataFrame({
        "Date": pd.date_range(start="2024-01-01", periods=NUM_DAYS),
        "Sales": np.random.randint(100, 501, size=NUM_DAYS),
        "Category": np.random.choice(["Electronics", "Clothing", "Home"], size=NUM_DAYS),
        "Profit": np.random.randint(10, 101, size=NUM_DAYS)
    })
    return df

if __name__ == "__main__":
    import timeit
    print("Original:", timeit.timeit(original, number=100))
    print("Optimized:", timeit.timeit(optimized, number=100))