Prev Next

Vectorized Operations

Vectorized operations in Pandas allow you to perform efficient computations on entire DataFrames or Series without explicit loops. These operations are highly optimized and take advantage of underlying NumPy arrays. This tutorial demonstrates common vectorized operations and their benefits.

Performing Elementwise Operations

You can perform arithmetic and other operations on entire columns or rows using vectorized operations. Here’s an example: 

import pandas as pd

# Create a sample DataFrame
data = {
    "A": [1, 2, 3],
    "B": [4, 5, 6]
}

df = pd.DataFrame(data)

# Perform elementwise addition
df["Sum"] = df["A"] + df["B"]
print(df)

Output:

A B Sum
1 4 5
2 5 7
3 6 9

Explanation: The operation df["A"] + df["B"] adds corresponding elements of columns A and B, creating a new column Sum.

Applying Universal Functions

Universal functions (ufuncs) like np.sqrt() or np.exp() can be applied directly to DataFrames or Series. Here’s an example: 

import numpy as np

# Apply square root to column A
df["Sqrt_A"] = np.sqrt(df["A"])
print(df)

Output:

A B Sum Sqrt_A
1 4 5 1.0
2 5 7 1.414
3 6 9 1.732

Explanation: The np.sqrt() function calculates the square root of each element in column A, creating a new column Sqrt_A.

Comparison and Logical Operations

Vectorized comparisons and logical operations can be applied to filter or transform data. Here’s an example: 

# Filter rows where A > 2
filtered_df = df[df["A"] > 2]
print(filtered_df)

Output:

A B Sum Sqrt_A
3 6 9 1.732

Explanation: The condition df["A"] > 2 filters rows where the value in column A is greater than 2, resulting in a subset of the DataFrame.

Key Takeaways

  • Efficient Computations: Vectorized operations perform elementwise computations without explicit loops.
  • Universal Functions: Use NumPy ufuncs for mathematical transformations.
  • Logical Operations: Apply vectorized comparisons and logical conditions for data filtering.
  • Performance: Vectorized operations leverage Pandas’ optimized internal architecture, ensuring faster performance than traditional loops.
Prev Next