reset_index()

noun

say it: “REE-set IN-deks” /ˌriˈsɛt ˈɪndɛks/

A pandas method that gives a table plain row numbers again from 0, either keeping the old row labels as a column or throwing them away.

What does reset_index() do in pandas?

reset_index() is a pandas method that numbers a table's rows 0, 1, 2 again. The old row labels, such as the groups a groupby() made, become an ordinary column; with drop=True they're thrown away instead.

import pandas as pd
orders = pd.DataFrame({"size": ["small", "large", "small"], "total": [12.0, 20.0, 14.0]})
by_size = orders.groupby("size")["total"].sum().reset_index()
print(by_size)
#     size  total
# 0  large   20.0
# 1  small   26.0

In a sentence

  • “Her top-ten list was numbered 0 to 9 after reset_index() with drop=True.”
  • “After the groupby, reset_index() turned the sizes back into an ordinary column.”

Level: Journeyman · More Python terms at this level

Updated