📚LearnHub

Course Lessons

Lesson 2 of 4

NumPy Fundamentals

Master NumPy arrays, vectorized operations, and mathematical functions for efficient numerical computing in Python.

30 minutes

NumPy Fundamentals

NumPy is the foundation of scientific computing in Python, providing powerful array operations and mathematical functions.

What is NumPy?

  • N-dimensional arrays: Efficient storage and operations
  • Vectorization: Fast operations without loops
  • Mathematical functions: Comprehensive library
  • Foundation: Base for Pandas, Scikit-learn, and more

Creating Arrays

NumPy arrays are more efficient than Python lists:

import numpy as np

arr = np.array([1, 2, 3, 4, 5])
zeros = np.zeros((3, 4))  # 3x4 array of zeros
ones = np.ones((2, 3))    # 2x3 array of ones
range_arr = np.arange(0, 10, 2)  # [0, 2, 4, 6, 8]

Array Properties

Understand your arrays:

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

Array Operations

Vectorized operations are fast and clean:

  • Element-wise arithmetic: +, -, *, /
  • Broadcasting: Operations on arrays of different shapes
  • Aggregations: sum(), mean(), std(), min(), max()
  • Universal functions: sqrt(), exp(), log(), sin()

Indexing and Slicing

Access and modify array elements:

arr[0]        # First element
arr[-1]       # Last element
arr[1:4]      # Slice
arr[arr > 5]  # Boolean indexing

Array Reshaping

Change array dimensions:

  • reshape(): Change shape without copying data
  • flatten(): Convert to 1D array
  • transpose(): Swap axes

Best Practices

  • Use vectorized operations instead of loops
  • Choose appropriate data types for memory efficiency
  • Leverage broadcasting for cleaner code
  • Pre-allocate arrays when size is known

Code Example

import numpy as np

# Creating arrays
temperatures = np.array([72, 75, 68, 71, 73, 69, 70])
print(f"Shape: {temperatures.shape}, Type: {temperatures.dtype}")

# Array operations (vectorized)
celsius = (temperatures - 32) * 5/9
print(f"Celsius: {celsius.round(1)}")

# Statistical operations
print(f"Mean: {temperatures.mean():.1f}°F")
print(f"Std Dev: {temperatures.std():.1f}°F")
print(f"Range: {temperatures.min()}°F - {temperatures.max()}°F")

# Boolean indexing
hot_days = temperatures[temperatures > 72]
print(f"Hot days (>72°F): {hot_days}")

# 2D arrays (matrices)
sales_data = np.array([
    [120, 130, 125],  # Week 1
    [115, 140, 135],  # Week 2
    [130, 145, 150],  # Week 3
    [125, 138, 142]   # Week 4
])

print(f"Sales shape: {sales_data.shape}")  # (4, 3)
print(f"Total sales: ${sales_data.sum()}")
print(f"Average per week: {sales_data.mean(axis=1).round(1)}")
print(f"Average per day: {sales_data.mean(axis=0).round(1)}")

# Array manipulation
flat_sales = sales_data.flatten()
reshaped = sales_data.reshape(6, 2)
print(f"Reshaped: {reshaped.shape}")

# Mathematical operations
prices = np.array([10.99, 15.50, 8.75])
quantities = np.array([5, 3, 7])
revenue = prices * quantities
print(f"Revenue per product: ${revenue}")
print(f"Total revenue: ${revenue.sum():.2f}")
Previous LessonNext Lesson