pandas.DataFrame(dtype=”category”) : For creating a categorical dataframe, dataframe() method has dtype attribute set to category.
All the columns in data-frame can be converted to categorical either during or after construction by specifying dtype=”category” in the DataFrame constructor.
Code :
# Python code explaining # constructing categorical data frame # importing libraries import numpy as np import pandas as pd # Constructing dataframe data = {'col1': [1, 2, 4, 5], 'col2': [3, 4, 5, 6]} df1 = pd.DataFrame(data = data) print ("df1 : \n", df1) print("\n\ndf1 type :\n", df1.dtypes) |
Output :
# Converting dataframe to category df2 = pd.DataFrame({'A': list('1245'), 'B': list('3456')}, dtype ="category") print ("df2 : \n", df2) print("\n\ndf2 type :\n", df2.dtypes) print ("\n\ndf2 column 0 :\n", df2['A']) print ("\n\ndf2 column 1 :\n", df2['B']) |
Output :
# Conversion can be done using astype() df3 = pd.DataFrame({'A': list('efgh'), 'B': list('aebc')}) print ("\n\ndf3 : \n", df3) print("\ndf3 type :\n", df3.dtypes) df4 = df3.astype('category') print ("\n\ndf4 type:\n", df4.dtypes) |
Output :
