pivot_table()

noun

say it: “PIV-ut TAY-bul” /ˈpɪvət ˈteɪbəl/

A pandas method that builds a grid summary from a table: the rows from one column, the columns from another, and a total or average in each cell.

What does pivot_table() do in pandas?

pivot_table() is the pandas method that builds a pivot table: index names the column down the side, columns the one across the top, values the numbers to summarize and aggfunc how (a sum, a mean, a count). fill_value fills empty squares and margins=True adds the totals.

import pandas as pd
orders = pd.DataFrame({"weekday": ["Fri", "Fri", "Sat"], "size": ["small", "large", "large"], "total": [12.0, 20.0, 22.0]})
grid = orders.pivot_table(index="weekday", columns="size", values="total",
                          aggfunc="sum", fill_value=0, margins=True)
print(grid)
# size     large  small   All
# weekday
# Fri       20.0   12.0  32.0
# Sat       22.0    0.0  22.0
# All       42.0   12.0  54.0

In a sentence

  • “She filled the empty squares with zeros through the fill_value of pivot_table().”
  • “pivot_table() with margins=True added a total row and a total column.”

Level: Journeyman · More Python terms at this level

Updated