pandas DataFrame nunique

pandas DataFrame.nunique function

DataFrame.nunique (axis=0, dropna=True)[source] Counts number of distinct elements in specified axis. Returns Series with number of distinct elements. Can ignore NaN values.

Name	Description	Type/Default Value	Required / Optional
axis	The axis to use. 0 or ‘index’ for row-wise, 1 or ‘columns’ for column-wise.	{0 or ‘index’, 1 or ‘columns’} Default Value: 0	Required
dropna	Don’t include NaN in the counts.	bool Default Value: True	Required

Python is a great language for data analysis, mainly because of the fantastic ecosystem of data-centric Python packages. Pandas is one of those packages and it makes importing and analyzing data a lot easier. The Pandas dataframe.nunique () function returns the series with the number of distinct observations on the requested axis. If we set the axis value to 0, it finds the total number of unique observations on the index axis. If we set the axis value to 1, it finds the total number of unique observations on the column axis. It also provides the functionality to exclude NaN values ​​from the unique number count.

pandas DataFrame nunique Example #1

def get_nunique(self, colname):
        """
        Looks up or caches the number of unique (distinct) values in a column,
        or calculates and caches it.
        """
        return self.get_cached_value(’nunique’, colname, self.calc_nunique) 

pandas DataFrame nunique Example #2

def get_database_nunique(self, tablename, colname):
        colname = self.quoted(colname)
        sql = (’SELECT COUNT(DISTINCT %s) FROM %s WHERE %s IS NOT NULL’
               % (colname, tablename, colname))
        return self.execute_scalar(sql) 

pandas DataFrame nunique Example #3

Use nunique() function to find the number of unique values over the column axis.

# importing pandas as pd
import pandas as pd
  
# Creating the first dataframe 
df = pd.DataFrame({"A":[14, 4, 5, 4, 1],
                   "B":[5, 2, 54, 3, 2],
                   "C":[20, 20, 7, 3, 8],
                    "D":[14, 3, 6, 2, 6]})
  
# Print the dataframe
df

pandas DataFrame nunique Example #4

Use nunique() function to find the number of unique values over the index axis in a dataframe. The dataframe contains NaN values

# importing pandas as pd
import pandas as pd

# Creating the first dataframe
df = pd.DataFrame({"A":["Sandy", "alex", "brook", "kelly", np.nan],
				"B":[np.nan, "olivia", "olivia", "", "amanda"],
				"C":[20 + 5j, 20 + 5j, 7, None, 8],
				"D":[14.8, 3, None, 6, 6]})

# apply the nunique() function
df.nunique(axis = 0, dropna = True)

Archived version

The Pandas function dataframe.nunique() returns a series with the number of different observations along the requested axis. If we set the axis value to 0, then it will find the total number of unique observations along the index axis. If we set the axis value to 1, we get the total number of unique observations along the column axis. It also provides a function to exclude NaN values ​​from unique numbers.

Syntax: DataFrame.nunique (axis = 0, dropna = True)

Parameters:
axis: {0 or ’index’, 1 or ’columns’}, default 0
dropna: Don’t include NaN in the counts.

Returns: nunique: Series

Example # 1: Use nunique () to find the number of unique values ​​along the column axis.

# import pandas as pd import pandas as pd   # Create first data frame df = pd.DataFrame ( { "A" : [ 14 , 4 , 5 , 4 , 1 ], "B" : [ 5 , 2 , 54 , 3 , 2 ], "C" : [ 20 , 20 , 7 , 3 , 8 ], "D" : [ 14 , 3 , 6 , 2 , 6 ]})   # Print the data frame df

Let’s use the dataframe.nunique () function to find unique values ​​along the column axis.

# find unique values ​​ df.nunique (axis = 1 )

Output:

As we can see in the output, the function prints the total number. unique values ​​in each row.

Example # 2: Use nunique () to find the number of unique values ​​along the index axis in a data frame. The data frame contains NaN values.

# import pandas as pd import pandas as pd   # Create first data frame df = pd.DataFrame ({ "A" : [ " Sandy " , " alex " , "brook" , "kelly" , np.nan],   " B " : [np.nan, "olivia" , "olivia" , " ", " amanda "],    " C " : [ 20 + 5j , 20 + 5j , 7 , None , 8 ], "D" : [ 14.8 , 3 , None , 6 , 6 ]})    # apply nunique () function df.nunique (axis = 0 , dropna = True )

Output:

The function treats an empty string as a unique value in column 2.