Python Basics

This section covers essential Python programming concepts for machine learning (ML) tasks. You’ll learn syntax, data types, and libraries, preparing for ML labs with numpy, pandas, and scikit-learn. Basic familiarity with Python helps, but no ML experience is required. Write the code yourself to learn Python, but you can refer to examples in our GitHub repository: https://github.com/ravishankarkumar/aiunderthehood-sample-code.

Why Python for ML?

Python dominates AI/ML because of its simplicity, readability, and massive ecosystem. Its high-level syntax enables rapid prototyping, while libraries like numpy, pandas, and pytorch handle performance-critical operations in optimized C/C++ under the hood.

Basic Syntax

Python’s syntax is simple and beginner-friendly. This program sums a list, showing variables, loops, and functions:

def main():
    numbers = [1, 2, 3, 4, 5]
    total = 0
    for num in numbers:
        total += num
    print("Sum:", total)

if __name__ == "__main__":
    main()

• Variables: Declared dynamically, no type annotations needed (total = 0).
• Lists: [] stores ordered data, commonly used for datasets.
• Loops: for iterates directly over items.
• Functions: def defines functions, with main() as convention.

Run with python main.py to see “Sum: 15”.

Data Structures for ML

ML tasks require arrays, matrices, and datasets. Python provides:

• Lists (list): Flexible collections for data.
• Tuples (tuple): Immutable sequences.
• Dictionaries (dict): Key-value pairs for metadata.
• NumPy Arrays: Efficient multidimensional arrays (preferred for ML).

Example dataset with Python structures:

dataset = {
    "features": [[1.0, 2.0], [3.0, 4.0]],
    "labels": [0, 1]
}

print("Dataset size:", len(dataset["features"]))

This mimics ML datasets, later used with numpy, pandas, or scikit-learn.

Lab: Vector Operations with NumPy

Practice Python by computing the Euclidean distance between two vectors, a key ML operation.

1. Create vector_distance.py in your project:

   import numpy as np
   
   def euclidean_distance(v1, v2):
       v1, v2 = np.array(v1), np.array(v2)
       if v1.shape != v2.shape:
           raise ValueError("Vectors must have equal length")
       return np.sqrt(np.sum((v1 - v2) ** 2))
   
   if __name__ == "__main__":
       v1 = [1.0, 2.0, 3.0]
       v2 = [4.0, 5.0, 6.0]
       distance = euclidean_distance(v1, v2)
       print(f"Euclidean Distance: {distance:.3f}")

2. Dependencies: Install NumPy if not already installed:

   pip install numpy

3. Run:

   python vector_distance.py

   Expected Output:

   Euclidean Distance: 5.196

This builds skills for ML computations, using NumPy for efficient vector operations.

Learning from Official Resources

Deepen your Python knowledge with:

• Python Official Tutorial: docs.python.org/tutorial
• Automate the Boring Stuff with Python by Al Sweigart: Beginner-friendly, practical examples.
• Effective Python by Brett Slatkin: Best practices for writing clean, efficient code.

Next Steps

Move to First ML Lab to build your first ML model, or revisit Setup.

Further Reading

• Python Documentation: docs.python.org
• Automate the Boring Stuff with Python by Al Sweigart
• Effective Python by Brett Slatkin