In this article, we are going to learn about String Munging In Pandas Dataframe. Munging is known as cleaning up anything which was messy by transforming them. In technical terms, we can say that transforming the data in the database into a useful form.
Example: “no-one@example.com”, becomes “no-one at example dot com”
Approach:
Step 1: import the library
Python3
import pandas as pd import numpy as np import re as re |
Step 2: creating Dataframe
Now create a dictionary and pass it through pd.DataFrame to create a Dataframe.
Python3
raw_data = {"first_name": ["Jason", "Molly", "Tina", "Jake", "Amy"], "last_name": ["Miller", "Jacobson", "Ali", "Milner", "Cooze"], "email": ["jas203@gmail.com", "momomolly@gmail.com", np.NAN, "battler@milner.com", "Ames1234@yahoo.com"]} df = pd.DataFrame(raw_data, columns=["first_name", "last_name", "email"]) print() print(df) |
Step 3: Applying Different Munging Operation
First, check that in feature “email” which string contains “Gmail”.
Python3
print(df["email"].str.contains("gmail")) |
Now we want to separate the email into parts such that characters before “@” becomes one string and after and before “.” becomes one. At last, the remaining becomes the one string.
Python3
pattern = "([A-Z0-9._%+-]+)@([A-Z0-9.-]+)\.([A-Z]{2,4})"print(df["email"].str.findall(pattern, flags=re.IGNORECASE)) |
Below is the implementation:
Python3
def ProjectPro_Ex_136(): print() print('**How we can do string munging in Pandas**') # loading libraries import pandas as pd import numpy as np import re as re # Creating dataframe raw_data = {'first_name': ['Jason', 'Molly', 'Tina', 'Jake', 'Amy'], 'last_name': ['Miller', 'Jacobson', 'Ali', 'Milner', 'Cooze'], 'email': ['jas203@gmail.com', 'momomolly@gmail.com', np.NAN, 'battler@milner.com', 'Ames1234@yahoo.com']} df = pd.DataFrame(raw_data, columns=['first_name', 'last_name', 'email']) print() print(df) # Let us find Which string within the # email column contains ‘gmail’ print() print(df['email'].str.contains('gmail')) # Create a daily expression pattern that # breaks apart emails pattern = '([A-Z0-9._%+-]+)@([A-Z0-9.-]+)\\.([A-Z]{2,4})' # Find everything in df.email that contains # that pattern print() print(df['email'].str.findall(pattern, flags=re.IGNORECASE)) ProjectPro_Ex_136() |
Output:
