Prev Next

Replacing Values in Pandas

The replace() method in Pandas provides a versatile way to substitute values in a DataFrame or Series. This method can handle single values, lists, and even dictionaries for more complex replacements. This tutorial demonstrates the key use cases for replace().

Replacing Single Values

Replace a specific value in a column or across the entire DataFrame. Here’s an example: 

import pandas as pd

# Create a sample DataFrame
data = {
    "Name": ["Karthick", "Durai", "Praveen"],
    "City": ["Chennai", "Coimbatore", "Chennai"]
}

df = pd.DataFrame(data)

# Replace 'Chennai' with 'Chennai Metro'
df["City"] = df["City"].replace("Chennai", "Chennai Metro")
print(df)

Output:

Name City
Karthick Chennai Metro
Durai Coimbatore
Praveen Chennai Metro

Explanation: The replace() method substitutes all occurrences of 'Chennai' with 'Chennai Metro' in the City column.

Replacing Multiple Values

Replace multiple values using a list or dictionary. Here’s an example: 

# Replace multiple values using a dictionary
replacement_dict = {
    "Chennai Metro": "Chennai",
    "Coimbatore": "Kovai"
}
df["City"] = df["City"].replace(replacement_dict)
print(df)

Output:

Name City
Karthick Chennai
Durai Kovai
Praveen Chennai

Explanation: The dictionary in replace() maps values to their replacements, converting 'Chennai Metro' to 'Chennai' and 'Coimbatore' to 'Kovai'.

Replacing Missing Values

Use replace() to handle missing values (e.g., NaN) in a DataFrame. Here’s an example: 

import numpy as np

# Add missing values
df.loc[1, "City"] = np.nan

# Replace NaN with 'Unknown'
df["City"] = df["City"].replace(np.nan, "Unknown")
print(df)

Output:

Name City
Karthick Chennai
Durai Unknown
Praveen Chennai

Explanation: The replace() method substitutes NaN values with 'Unknown' in the City column, ensuring no missing data remains.

Key Takeaways

  • Versatility: The replace() method handles single values, multiple values, and missing data efficiently.
  • Customization: Use dictionaries for complex replacements and lists for bulk replacements.
  • Missing Data: Replace NaN values to ensure clean and complete datasets.
  • Scalability: replace() works seamlessly on large datasets with diverse replacement needs.
Prev Next