NumPy for Numerical Computing

NumPy is a powerful Python library used for numerical computing. It provides efficient data structures and functions for handling large datasets, performing mathematical operations, and supporting deep learning workflows. NumPy is a foundational tool for AI and data science because it enables fast computation with arrays and matrices.

Why NumPy is Important
Traditional Python lists are flexible but slow for large-scale numerical computations. NumPy arrays are more efficient in memory usage and computation speed. They support vectorized operations, which allow performing calculations on entire datasets without explicit loops, making your code faster and cleaner.

NumPy Arrays
The core of NumPy is the ndarray, a multi-dimensional array object:

import numpy as np# Create a 1D array
arr1 = np.array([1, 2, 3, 4])# Create a 2D array
arr2 = np.array([[1, 2], [3, 4]])

Array Attributes
NumPy arrays have attributes that provide useful information:

• shape – Dimensions of the array
• dtype – Data type of the elements
• size – Total number of elements
• ndim – Number of dimensions

Array Operations
NumPy allows fast element-wise operations:

a = np.array([1, 2, 3])
b = np.array([4, 5, 6])# Element-wise addition
c = a + b# Element-wise multiplication
d = a * b

Mathematical Functions
NumPy provides a variety of mathematical functions for arrays:

• np.sum() – Sum of elements
• np.mean() – Mean of elements
• np.max() / np.min() – Maximum and minimum values
• np.sqrt() – Square root
• np.exp() – Exponential function

Indexing and Slicing
You can access and modify array elements using indexing and slicing:

arr = np.array([10, 20, 30, 40, 50])# Access first element
print(arr[0])# Slice elements
print(arr[1:4])

Reshaping and Broadcasting
NumPy supports reshaping arrays and broadcasting to perform operations on arrays of different shapes:

arr = np.arange(6)
arr = arr.reshape(2, 3)  # 2 rows, 3 columns

Broadcasting allows arithmetic between arrays of different shapes without explicit replication.

Applications in Deep Learning

• Representing inputs, weights, and outputs as arrays or tensors
• Performing matrix multiplications for neural network computations
• Data preprocessing, normalization, and scaling
• Fast numerical operations in training deep learning models

Lesson Summary
In this lesson, you learned the basics of NumPy, including arrays, attributes, operations, indexing, reshaping, and its applications in deep learning. Mastering NumPy is essential for efficient numerical computing and building AI models.