Pandas rank() Function

Examples of the rank() Function in Pandas

Example 1

For the first example, we will create a dataframe containing two columns, product_name and quantity.

• Python

Python

import pandas as pd
 

data = {'product_name': ['Apple', 'Banana', 'Orange', 'Grapes', 'Kiwi'],

        'quantity': [10, 15, 8, None, 15]}
 

df = pd.DataFrame(data)
 

print(df)

You can also try this code with Online Python Compiler

Run Code

Output:

Now, we will apply the rank function with varying parameters.

• Python

Python

print(df)

# Applying rank() with default parameters

df['Rank_Default'] = df['quantity'].rank()
 

# Applying rank() with method='min' and ascending=False

df['Rank_MethodMin_Desc'] = df['quantity'].rank(method='min')
 

# Applying rank() with method='max'

df['Rank_MethodMax'] = df['quantity'].rank(method='max')
 

# Applying rank() with method='first'

df['Rank_MethodFirst'] = df['quantity'].rank(method='first')
 

# Applying rank() with method='dense'

df['Rank_MethodDense'] = df['quantity'].rank(method='dense')



# Applying rank() with na_option='top'

df['Rank_NA_Top'] = df['quantity'].rank(na_option='top')

df.head()

You can also try this code with Online Python Compiler

Run Code

Output:

Here, we call the rank function on the quantity column and add the returned series as additional columns to the dataframe.

Example 2

We applied the rank() function to a single column in the previous example. Now let’s see how to use it on an entire dataframe.

• Python

Python

import pandas as pd
 

data = {'product_name': ['Apple', 'Banana', 'Orange', 'Grapes'],

        'quantity': [10, 15, 8, 20],

        'price': [0.5, 0.25, 0.3, 0.6]}



df = pd.DataFrame(data)



ranked_df = df.rank(numeric_only=True)
 

print("Original DataFrame:")

print(df)
 

print("\nDataFrame with Rank Values:")

print(ranked_df)

You can also try this code with Online Python Compiler

Run Code

Output:

Here, the result is a dataframe because we called the rank() function on an entire dataframe instead of a column, and as the axis parameter is 0 by default, the rank is generated along rows.

Example 3

In this example, we will set the pct parameter to true.

• Python

Python

import pandas as pd
 

data = {'Product': ['Apple', 'Banana', 'Orange', 'Grapes', 'Pear'],

        'Sales': [100, 150, 200, 150, 200]}
 

df = pd.DataFrame(data)

print("Original DataFrame:")

print(df)

print("\n")
 

df['Rank_Pct'] = df['Sales'].rank(pct=True)
 

print(df)

You can also try this code with Online Python Compiler

Run Code

Output:

Here, the rankings are displayed in percentile form.

In the following examples, we will use the rank() function in combination with functions such as groupby() and sort_values().

Example 4

In this example, we will use the rank() function to calculate ranks within groups in our dataframe. First, we will create a dataframe with 3 columns - Category, Product, and Quantity and then we will use the groupby() function to perform grouping based on the Category column. After the grouping, we will use the rank() function on the Quantity column with default parameters.

• Python

Python

import pandas as pd
 

data = {'Category': ['Fruit', 'Fruit', 'Vegetable', 'Fruit', 'Vegetable', 'Fruit'],

        'Product': ['Apple', 'Banana', 'Carrot', 'Grapes', 'Cabbage', 'Orange'],

        'Quantity': [10, 15, 8, 20, 5, 12]}
 

df = pd.DataFrame(data)

print("Original DataFrame:")

print(df)

print("\n")
 

# Applying rank() within each group using groupby()

df['Rank_Within_Group'] = df.groupby('Category')['Quantity'].rank()



print(df)

You can also try this code with Online Python Compiler

Run Code

Output:

Here, row indices 0 and 4 are ranked 1 because they have the lowest quantity in their respective groups.

Example 5

This example has a dataframe with 3 columns - Name, Math, and Science. We will first calculate the ranks separately based on the Math and Science columns, and then we will find the mean rank for each student and sort the dataframe based on the mean rank.

• Python

Python

import pandas as pd
 

data = {'Name': ['Alice', 'Bob', 'Charlie', 'David'],

        'Math': [85, 92, 78, 92],

        'Science': [70, 88, 95, 82]}
 

df = pd.DataFrame(data)



# Calculate ranks for 'Math' and 'Science'

df['Math_Rank'] = df['Math'].rank(ascending=False, method='min')

df['Science_Rank'] = df['Science'].rank(ascending=False, method='min')
 

# Calculate the mean rank for each student

df['Mean_Rank'] = df[['Math_Rank', 'Science_Rank']].mean(axis=1)
 

# Sort according to the mean rank

df_sorted = df.sort_values(by='Mean_Rank')
 

print("Original DataFrame:")

print(df)
 

print("\nDataFrame Sorted by Mean Rank:")

print(df_sorted)

You can also try this code with Online Python Compiler

Run Code

Output:

Note that the ascending parameter was false, and the method parameter was min for breaking ties. The student with the highest scores gets a rank closer to 1.

Frequently Asked Questions

Does the rank() function in Pandas modify the original data?

The rank() function doesn't modify the original Dataframe or Series. It returns a new Dataframe or Series containing the calculated rank values based on the input data structure.

What is NaN in Python?

In Python, NaN stands for “Not a Number, " a special floating point value used to represent undefined or unrepresentable numerical values. In Pandas, NaN represents missing values that arise from concatenation operations performed on dataframes.

How can I use the rank() function to handle missing values?

The rank() function allows you to control the treatment of missing values using the na_option parameter, which can be set to - keep, top, or bottom.

Conclusion

In this article, you learned about the rank() function in Pandas with the help of various examples. We also used it in combination with other functions provided by Pandas, such as groupby() and mean().

Go through the following articles to learn more about Pandas:-

• Getting Started with Pandas
   
• Pandas Pivot Table
   
• Advanced functions in Pandas
   

Happy Learning!