Vectorized Operation

noun

say it: “VEK-tuh-ryzd op-uh-RAY-shun” /ˈvɛktəˌraɪzd ˌɑpəˈreɪʃən/

Math written once for a whole column or array, such as price times quantity, that runs on every value at once with no loop.

What is a vectorized operation?

A vectorized operation is math done on a whole array or column at once, such as prices * 0.9, with no loop of your own. NumPy runs the loop inside itself, which is much quicker than a Python for loop.

import pandas as pd
orders = pd.DataFrame({"price": [4.50, 6.00], "quantity": [2, 3]})
orders["total"] = orders["price"] * orders["quantity"]   # every row at once
print(orders["total"].tolist())   # [9.0, 18.0]

In a sentence

  • “The total column came from one vectorized operation: price times quantity.”
  • “Swapping the loop for a vectorized operation made the script finish in a second.”

Level: Journeyman · More Python terms at this level

Updated