Skip to content
LearnPY
pandasfunction reference

pd.DataFrame

Build a two-dimensional labeled table.

Signature

pd.DataFrame(data=None, index=None, columns=None)

What it does

A DataFrame stores columns that can have different dtypes while sharing one row index. You can construct one from a dictionary of columns, a list of records, or another table.

Parameters

datadict, list, ndarray, or Series, optionaloptional
The values that become the table.
indexIndex or array_like, optionaloptional
Row labels. A default RangeIndex is used if omitted.
columnsIndex or array_like, optionaloptional
Column labels, useful when data has no names.

Returns

DataFrame

Example

pythonpd.DataFrame example
import pandas as pd

df = pd.DataFrame({
    "team": ["red", "blue", "red"],
    "score": [10, 12, 8],
})
print(df)

Output

   team  score
0   red     10
1  blue     12
2   red      8

Watch out for

  • A dictionary of scalars needs an index; without one, pandas raises a ValueError.
  • Column names are labels, not positions; use .iloc when you mean positions.