Sorting and ranking in Python

Sort by values or by index, and rank without reordering.

import pandas as pd

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

print(frame.sort_values(["tpm", "stars"], ascending=[False, True]))
print(frame.nlargest(2, "tpm"))
print(frame["tpm"].rank(method="min", ascending=False).tolist())

How it works

1. sort_values takes one column or a list, with ascending per column.
2. nlargest is the cheap way to ask for a top ten.
3. rank numbers the rows in place, ties averaged by default.

Keywords and builtins used here

• as
• print

The run, in numbers

Lines

11

Characters to type

304

Tokens

127

Three-star pace

105 tpm

At the three-star pace of 105 tokens a minute, this run takes about 73 seconds.

Type this snippet

Step 4 of 4 in Selecting rows, step 8 of 26 in pandas.

← Previous Next →