groupby in Python

Split by a key, apply a function, combine the answers.

import pandas as pd

frame = pd.DataFrame({
    "lang": ["python", "python", "rust", "rust", "sql"],
    "stars": [3, 1, 3, 2, 2],
    "tpm": [104, 91, 98, 88, 79],
})

print(frame.groupby("lang")["tpm"].mean().round(1))
print(frame.groupby("lang").size())
print(frame.groupby("lang", as_index=False)["stars"].max())

How it works

1. A single aggregate over a groupby gives one row per group.
2. as_index=False keeps the key as a column instead of the index.
3. size counts rows, nunique counts distinct values.

Keywords and builtins used here

• as
• print

The run, in numbers

Lines

11

Characters to type

304

Tokens

135

Three-star pace

110 tpm

At the three-star pace of 110 tokens a minute, this run takes about 74 seconds.

Type this snippet

Step 1 of 5 in Grouping, step 14 of 26 in pandas.

← Previous Next →