How to return a subset of a DataFrame's columns based on dtypes

Overview

The select_dtypes() function in pandas returns a subset of a given DataFrame's columns based on the column data types.

Syntax

The select_dtypes() function's syntax is shown below:

DataFrame.select_dtypes(include=None, exclude=None)
Syntax for the select_dtypes() function

Parameter value

The select_dtypes() function takes any of the two parameter values: include and exclude (at least one of them must be supplied). These represent the selection of the data type(s) to be included or excluded.

Return value

The select_dtypes() function returns a subset of the DataFrame that includes the data types in include and excludes the data types in exclude.

Example

# A code to illustrate the select_dtypes() function in Pandas



# importing the pandas library

from pandas import DataFrame



# creating a dataframe

my_data_frame = DataFrame({'Id': [1, 2, 3, 4, 5, 6],

                   'Married': [True, False] * 3,

                   'Score': [1.0, 2.0, 3.0, 4.0, 5.0, 6.0]})



# obtaining the columns by their data types

bool_data = my_data_frame.select_dtypes(include=bool)

int_and_float_data = my_data_frame.select_dtypes(include=[int, float])



# obtaining the columns by excluding data types

data_without_int_values = my_data_frame.select_dtypes(exclude=int)



# printing results

print(bool_data)

print(int_and_float_data)

print(data_without_int_values)

Explanation

  • Line 4: We import the DataFrame from the pandas library.
  • Lines 7–9: We create a DataFrame object, my_data_frame.
  • Lines 12–16: We obtain the columns of the DataFrame based on their data types using the select_dtypes() function. We assign the results to the bool_data, int_and_float_data and data_without_int_values variables.
  • Lines 19–21: We print the values of bool_data, int_and_float_data and data_without_int_values.

Free Resources

License: Creative Commons-Attribution-ShareAlike 4.0 (CC-BY-SA 4.0)