Python Pandas - Return a new Index of the values set with the mask

To return a new Index of the values set with the mask, use the index.putmask() method in Pandas. This method creates a new Index where values meeting a specified condition are replaced with a new value.

Syntax

Index.putmask(mask, value)

Parameters

The putmask() method accepts the following parameters:

• mask ? A boolean condition that determines which values to replace
• value ? The replacement value for positions where mask is True

Example

Let's create a Pandas Index and demonstrate how putmask() works ?

import pandas as pd

# Creating Pandas index
index = pd.Index([5, 65, 10, 17, 75, 40])

# Display the Pandas index
print("Original Index:")
print(index)

# Display index properties
print("\nNumber of elements:", index.size)
print("Data type:", index.dtype)

# Replace values less than 30 with 111
new_index = index.putmask(index < 30, 111)
print("\nAfter putmask (values < 30 replaced with 111):")
print(new_index)

Original Index:
Index([5, 65, 10, 17, 75, 40], dtype='int64')

Number of elements: 6
Data type: int64

After putmask (values < 30 replaced with 111):
Index([111, 65, 111, 111, 75, 40], dtype='int64')

Multiple Conditions Example

You can also use more complex boolean conditions with putmask() ?

import pandas as pd

# Create index with different values
index = pd.Index([1, 25, 50, 75, 100, 125])

print("Original Index:")
print(index)

# Replace values between 20 and 80 with -1
mask = (index >= 20) & (index <= 80)
new_index = index.putmask(mask, -1)

print("\nAfter putmask (values between 20-80 replaced with -1):")
print(new_index)

Original Index:
Index([1, 25, 50, 75, 100, 125], dtype='int64')

After putmask (values between 20-80 replaced with -1):
Index([1, -1, -1, -1, 100, 125], dtype='int64')

Key Points

• putmask() returns a new Index without modifying the original
• The mask parameter must be a boolean array or condition
• All values where mask is True get replaced with the specified value
• The returned Index maintains the same dtype as the original

Conclusion

The putmask() method provides an efficient way to conditionally replace values in a Pandas Index. Use it when you need to create a new Index with specific values replaced based on boolean conditions.