To rank the rows of Pandas DataFrame we can use the DataFrame.rank() method which returns a rank of every respective index of a series passed. The rank is returned on the basis of position after sorting.
Example #1 :
Here we will create a DataFrame of movies and rank them based on their ratings.
# import the required packages import pandas as pd # Define the dictionary for converting to dataframe movies = {'Name': ['The Godfather', 'Bird Box', 'Fight Club'], 'Year': ['1972', '2018', '1999'], 'Rating': ['9.2', '6.8', '8.8']} df = pd.DataFrame(movies) print(df) |
Output:
# Create a column Rating_Rank which contains # the rank of each movie based on rating df['Rating_Rank'] = df['Rating'].rank(ascending = 1) # Set the index to newly created column, Rating_Rank df = df.set_index('Rating_Rank') print(df) |
Output:
# Sort the dataFrame based on the index df = df.sort_index() print(df) |
Output:
Example #2
Let’s take an example of marks scored by 4 students. We will rank the students based on the highest mark they have scored.
# Create a dictionary with student details student_details = {'Name':['Raj', 'Raj', 'Raj', 'Aravind', 'Aravind', 'Aravind', 'John', 'John', 'John', 'Arjun', 'Arjun', 'Arjun'], 'Subject':['Maths', 'Physics', 'Chemistry', 'Maths', 'Physics', 'Chemistry', 'Maths', 'Physics', 'Chemistry', 'Maths', 'Physics', 'Chemistry'], 'Marks':[80, 90, 75, 60, 40, 60, 80, 55, 100, 90, 75, 70] } # Convert dictionary to a DataFrame df = pd.DataFrame(student_details) print(df) |
Output:
# Create a new column with Marks # ranked in descending order df['Mark_Rank'] = df['Marks'].rank(ascending = 0) # Set index to newly created column df = df.set_index('Mark_Rank') print(df) |
Output:
# Sort the DataFrame based on the index df = df.sort_index() print(df) |
Output:
Explanation:
Notice here that we have Raj and Arjun getting 90 marks each and hence they get ranked 2.5 (average of 2nd and 3rd rank i.e the two ranks they share). This can be seen for other marks in the table as well.
