Pandas DataFrame convert_dtypes() Method

Example

Convert the data types to better fit the content:

import pandas as pd

data = {
"name": ["Sally", "Mary", pd.NA],
"qualified": [True, False, pd.NA]
}

df = pd.DataFrame(data)

print("Original dtypes:")
print(df.dtypes)

newdf = df.convert_dtypes()

print("New dtypes:")
print(newdf.dtypes)

Try it Yourself »

Definition and Usage

The convert_dtypes() method returns a new DataFrame where each column has been changed to the best possible data type.

Syntax

dataframe.convert_dtypes(infer_objects, convert_string, convert_integer, convert_boolean, convert_floating)

Parameters

The parameters are keyword arguments.

Parameter	Value	Description
infer_objects	True\|False	Optional. Default True. Specifies whether to convert object dtypes to the best possible dtype or not.
convert_string	True\|False	Optional. Default True. Specifies whether to convert object dtypes to strings or not.
convert_integer	True\|False	Optional. Default True. Specifies whether to convert object dtypes to integers or not.
convert_boolean	True\|False	Optional. Default True. Specifies whether to convert object dtypes to booleans or not.
convert_floating	True\|False	Optional. Default True. Specifies whether to convert object dtypes to floating types or not.

Return Value

a Pandas DataFrame with the converted result.

❮ DataFrame Reference