Get unique values from a column in Pandas DataFrame

In Pandas, retrieving unique values from DataFrame is used for analyzing categorical data or identifying duplicates. Let's learn how to get unique values from a column in Pandas DataFrame.

Get the Unique Values of Pandas using unique()

The.unique()method returns a NumPy array. It is useful for identifying distinct values in a column, which can be helpful when working with categorical data or detecting outliers. The order of the unique values is preserved based on their first occurrence.

Syntax: DataFrame['column_name'].unique()

Consider the following example: we are retrieving and printing the unique values from the 'B' column using the unique() method.

# Import pandas package
import pandas as pd

# create a dictionary with five fields each
data = {
    'A': ['A1', 'A2', 'A3', 'A4', 'A5'],
    'B': ['B1', 'B2', 'B3', 'B4', 'B4'],
    'C': ['C1', 'C2', 'C3', 'C3', 'C3'],
    'D': ['D1', 'D2', 'D2', 'D2', 'D2'],
    'E': ['E1', 'E1', 'E1', 'E1', 'E1']}

# Convert the dictionary into DataFrame
df = pd.DataFrame(data)
print("Pandas DataFrame:")
display(df)

# Get the unique values of 'B' column
unique_values = df['B'].unique()

# Print the unique values
print("\nUnique values in 'B' column:")
print(unique_values)

Output:

The unique values returned are ['B1', 'B2', 'B3', 'B4'].

Find the unique values in a column using nunique()

Let's use .nunique() method to get the count of unique values in each column of the above dataframe.

# Get number of unique values in column 'A'
unique_values_A = df['A'].nunique()
# Print the number of unique values
print("Number of unique values in 'A' column:", unique_values_A)

# Get number of unique values in column 'B'
unique_values_B = df['B'].nunique()
# Print the number of unique values
print("\nNumber of unique values in 'B' column:", unique_values_B)

# Get number of unique values in column 'C'
unique_values_C = df['C'].nunique()
# Print the number of unique values
print("\nNumber of unique values in 'C' column:", unique_values_C)

# Get number of unique values in column 'D'
unique_values_D = df['D'].nunique()
# Print the number of unique values
print("\nNumber of unique values in 'D' column:", unique_values_D)

Output:

Number of unique values in 'A' column: 5
Number of unique values in 'B' column: 4
Number of unique values in 'C' column: 3
Number of unique values in 'D' column: 2

In addition to the .unique() method, there are other ways to retrieve unique values from a Pandas DataFrame, including:

Get Unique values from a Column in Pandas DataFrame using .drop_duplicates()

The .drop_duplicates() method removes duplicate values in the specified column, returning a DataFrame with only the unique values.

Syntax: DataFrame['column_name'].drop_duplicates()

Example: Get unique values from column 'C'

unique_values = df['C'].drop_duplicates()

print(unique_values)

Output:

0    C1
1    C2
2    C3
Name: C, dtype: object

This method returns the unique values as a Series and preserves the index of the original DataFrame.

Extracting Unique values in Pandas DataFrame Using .value_counts()

The .value_counts() method counts the occurrences of each unique value in the column and returns the result as a Series.

Syntax: DataFrame['column_name'].value_counts()

Example: Get unique values from column 'D' along with their counts

unique_values_count = df['D'].value_counts()

print(unique_values_count)

Output:

D
D2    4
D1    1
Name: count, dtype: int64

This method provides both the unique values and the frequency of each value. To extract just the unique values, you can use .index on the result.

unique_values = df['D'].value_counts().index

print(unique_values)

Output:

Index(['D2', 'D1'], dtype='object', name='D')

Get Unique values from a column in Pandas DataFrame using set()

You can also use Python’s built-in set() function, which converts the column values into a set, automatically removing duplicates.

Syntax: set(DataFrame['column_name'])

Example: Get unique values from column 'D'

unique_values = set(df['D'])

print(unique_values)

Output:

{'D1', 'D2'}

Using set() does not preserve the order of the unique values, but it is a quick way to get distinct values.

In short:

• The .unique() method returns a NumPy array of unique values, preserving their order of appearance.
• The .drop_duplicates() method returns a Series with unique values, preserving the original index.
• The .value_counts() method provides both the unique values and their frequency count.
• The set() function quickly returns unique values but does not preserve their order.