01-python

Module 01: Python for AI & Numerical Computing

Master the foundational language of Artificial Intelligence: Python 3, object-oriented principles, virtual environments, and high-performance array operations with NumPy.

---

📋 Prerequisites

• Basic computer literacy and logical thinking.
• No prior programming experience required.

---

🧠 Core Concepts

1. Python Fundamentals:
   • Variables, dynamic typing, primitive data types (int, float, str, bool).
   • Control flow: if-elif-else, while, and for loops.
   • Built-in data structures: Lists, Tuples, Sets, and Dictionaries.
   • Functions, *args, **kwargs, lambda expressions, and scope.
2. Object-Oriented Programming (OOP) for ML:
   • Classes, attributes, methods, __init__, and __call__ (crucial for custom PyTorch modules).
   • Inheritance and polymorphism.
3. Environment Management:
   • Why virtual environments matter.
   • Using python -m venv to isolate dependencies.
   • Installing and freezing requirements (pip install, pip freeze > requirements.txt).
4. Numerical Computing with NumPy:
   • Multidimensional arrays (ndarray), shape, dimensions, and data types (dtype).
   • Slicing and fancy indexing.
   • Vectorization vs. slow Python loops.
   • Broadcasting rules.
   • Linear algebra functions (np.dot, @, np.linalg.inv, np.linalg.eig).

---

🗺️ Recommended Sequence

1. Read The Python Tutorial (Chapters 1 to 5).
2. Set up a local virtual environment in VS Code.
3. Practice writing modular functions and custom classes.
4. Read the NumPy Quickstart Tutorial.
5. Complete the practical exercises below.

---

💻 Practical Exercises

Exercise 1: Vectorized Euclidean Distance

Implement Euclidean distance between two vectors using NumPy without using any for loops:

import numpy as np

def euclidean_distance(v1: np.ndarray, v2: np.ndarray) -> float:
    """Calculate Euclidean distance using vectorized NumPy operations."""
    diff = v1 - v2
    return np.sqrt(np.sum(diff ** 2))

# Test
a = np.array([1.0, 2.0, 3.0])
b = np.array([4.0, 6.0, 8.0])
print("Distance:", euclidean_distance(a, b))  # Expected ~6.403

Exercise 2: Matrix Multiplication Benchmark

Benchmark the performance difference between a 3-loop native Python matrix multiplication vs. np.dot / @ on two $200 \times 200$ matrices. Observe the $50\times - 100\times$ speedup from vectorization!

---

💡 Project Ideas

• Statistical CLI Tool: A command-line Python script that reads any CSV file, parses numbers, and computes mean, median, standard deviation, and IQR using NumPy.
• Custom Array Class: Build an educational mini-array class from scratch with basic element-wise addition, subtraction, and dot product.

---

📖 Official Documentation & Resources

• 📖 Official Python 3 Documentation
• 📖 NumPy User Guide
• 🎓 CS50’s Introduction to Programming with Python