NumPy Eigenvectors Explained – Python np.linalg.eig() with Examples and Visualization

NumPy – Eigenvectors 

Eigenvectors are a fundamental concept in linear algebra used extensively in:

• Machine learning
• Data science
• Physics
• Computer graphics
• Signal processing

In Python, NumPy makes it easy to compute eigenvectors using np.linalg.eig().

---

What is an Eigenvector?

An eigenvector is a special vector that does not change direction when a matrix transformation is applied.

Only its magnitude (length) changes.

---

Eigenvector Equation

The mathematical form is:

A v = λ v

Where:

• A = matrix
• v = eigenvector
• λ = eigenvalue

---

Simple Meaning

Eigenvectors are directions that remain unchanged after transformation.

Only stretching or shrinking happens.

---

Import NumPy

import numpy as np

---

1. Eigenvectors and Eigenvalues

import numpy as np

A = np.array([[4, 2],
              [1, 3]])

eigenvalues, eigenvectors = np.linalg.eig(A)

print("Eigenvalues:\n", eigenvalues)
print("Eigenvectors:\n", eigenvectors)

---

Output:

Eigenvalues:
[5. 2.]

Eigenvectors:
[[ 0.89442719 -0.70710678]
 [ 0.4472136   0.70710678]]

---

Explanation

• Each column in eigenvectors matrix is an eigenvector
• Each eigenvector corresponds to one eigenvalue
• Direction remains fixed, only scaling changes

---

2. Accessing Individual Eigenvectors

import numpy as np

A = np.array([[4, 2],
              [1, 3]])

eigenvalues, eigenvectors = np.linalg.eig(A)

v1 = eigenvectors[:, 0]
v2 = eigenvectors[:, 1]

print("Eigenvector 1:", v1)
print("Eigenvector 2:", v2)

---

3. Verify Eigenvector Property

Check if:

A v = λ v

import numpy as np

A = np.array([[4, 2],
              [1, 3]])

eigenvalues, eigenvectors = np.linalg.eig(A)

v = eigenvectors[:, 0]
λ = eigenvalues[0]

print(np.allclose(A @ v, λ * v))

---

Output:

True

---

4. 3×3 Matrix Eigenvectors

import numpy as np

A = np.array([[1, 2, 3],
              [0, 1, 4],
              [5, 6, 0]])

eigenvalues, eigenvectors = np.linalg.eig(A)

print("Eigenvectors:\n", eigenvectors)

---

5. Real vs Complex Eigenvectors

import numpy as np

A = np.array([[0, -1],
              [1,  0]])

eigenvalues, eigenvectors = np.linalg.eig(A)

print("Eigenvalues:", eigenvalues)
print("Eigenvectors:\n", eigenvectors)

---

Output (complex values):

[0.+1.j 0.-1.j]

---

Properties of Eigenvectors

✔ Do not change direction
✔ Only scale changes
✔ Exist only for square matrices
✔ Can be real or complex

---

Real-World Applications

Eigenvectors are used in:

1. Machine Learning

• PCA (Principal Component Analysis)
• Feature reduction

2. Data Science

• Pattern detection
• Data compression

3. Physics

• Quantum states
• Vibration analysis

4. Computer Graphics

• Rotation systems
• Transformation stability

---

Eigenvectors vs Eigenvalues

Concept	Meaning
Eigenvector	Direction that stays unchanged
Eigenvalue	Scaling factor of that direction

---

Why Eigenvectors Matter

Eigenvectors help to:

• Understand data structure
• Reduce dimensions
• Identify hidden patterns
• Solve transformation problems

---

Common Error

❌ Non-square matrix

ValueError: Last 2 dimensions of the array must be square

Solution:

Eigenvectors only exist for square matrices.

---

Why Use NumPy?

Using NumPy allows:

• Fast eigen decomposition
• Efficient linear algebra computation
• Easy integration with AI/ML systems

Combined with Python, it becomes essential for scientific computing.

---

Summary

Eigenvectors represent stable directions in a matrix transformation, while eigenvalues represent scaling factors.

With NumPy, you can compute them easily using:

np.linalg.eig(A)

---

Conclusion

Eigenvectors are a core concept in linear algebra and are essential for AI, machine learning, and data science. NumPy makes these advanced mathematical operations simple and accessible.