NUMPY Tutorial

🎯 NumPy Fancy Indexing

Fancy indexing means indexing NumPy arrays using integer or boolean arrays rather than just simple slices or single indices. It allows you to select arbitrary elements in a flexible way.

✅ Indexing with Integer Arrays

You can pass a list or NumPy array of integers to select elements at those positions.

import numpy as np

arr = np.array([10, 20, 30, 40, 50])
indices = [1, 3, 4]
print(arr[indices])  # Output: [20 40 50]

🎲 Multi-dimensional Fancy Indexing

For 2D arrays, you can pass lists/arrays for rows and columns to pick elements at those coordinate pairs.

arr2d = np.array([[1, 2, 3], 
                  [4, 5, 6], 
                  [7, 8, 9]])

rows = [0, 1, 2]
cols = [2, 1, 0]

print(arr2d[rows, cols])  # Output: [3 5 7]
# Picks elements at (0,2), (1,1), and (2,0)

🧪 Boolean Fancy Indexing

You can also use boolean arrays to select elements where the condition is True.

arr = np.array([10, 15, 20, 25, 30])
mask = arr > 18
print(arr[mask])  # Output: [20 25 30]

⚠️ Fancy Indexing Creates Copies, Not Views

Unlike slicing, fancy indexing returns a copy of the data, so modifying the result does NOT affect the original array.

arr = np.array([10, 20, 30, 40, 50])
indices = [1, 3]
fancy = arr[indices]
fancy[0] = 999

print(arr)    # Original unchanged: [10 20 30 40 50]
print(fancy)  # Modified copy: [999  40]

🔀 Combining Fancy Indexing and Slicing

You can combine fancy indexing with slicing to select complex subsets.

arr2d = np.arange(16).reshape(4,4)
print(arr2d)

# Select rows 1 and 3, columns 0 to 2 (excluding 2)
print(arr2d[[1,3], 0:2])  
# Output:
# [[4 5]
#  [12 13]]

🔎 Quick Tips

• Integer array indexing picks elements at exact positions.
• Boolean indexing filters elements based on condition(s).
• Fancy indexing always returns a copy.
• Can be used for advanced selections and manipulations.