NumPy: Introduction and Usage Guide

:dart: Learning Objectives

By the end of this lesson, you will be able to:

• Understand what NumPy is and why we use it
• Create and work with NumPy arrays
• Access and filter data in arrays
• Perform mathematical operations on arrays
• Use NumPy for data analysis tasks

:information_source: What is NumPy? NumPy stands for Numerical Python. It's a special Python package that helps us work with numbers and arrays much faster than regular Python lists. Think of it as a super-powered calculator for handling lots of numbers at once!

Why Use NumPy Instead of Lists?

• Speed matters: Python lists are slow when working with lots of numbers
• Built for math: NumPy arrays work faster and make math operations easier
• Special powers: You can add, multiply, and do other math operations on entire arrays at once!

:star2: Why NumPy is Important

• Handle tables of data: Store data in rows and columns using 2D arrays (like a spreadsheet!)
• Foundation for data science: Many popular tools like Pandas and Matplotlib are built on NumPy
• Professional tool: Data scientists use NumPy every day for their work

:bulb: Want to explore more? Visit the official NumPy website: https://numpy.org/

:books: What We'll Learn Today

We'll explore NumPy through hands-on examples:

1. Install and Import NumPy - Set up NumPy on your computer
2. Create and Generate NumPy arrays - Make different types of arrays
3. NumPy array properties - Learn about array characteristics
4. Random in NumPy - Generate random numbers
5. Access Items in NumPy array - Get specific values from arrays
6. Filter Items in NumPy array - Find data that meets certain conditions
7. Mathematical operations on NumPy arrays - Do math with entire arrays

:wrench: Install and Import NumPy

Installation Steps

• VSCode users: Open your terminal and type:
  bash

  py -m pip install numpy
  

• Google Colab users: Good news! NumPy is already installed for you!

Import NumPy

After installation, import NumPy into your program:

python

import numpy as np

note We use np as a shortcut name for NumPy. This makes our code shorter and easier to read!

:hammer_and_wrench: Create and Generate NumPy Arrays

Creating Arrays from Lists

Let's start by creating a simple array with numbers from 1 to 5:

array = np.array([1,2,3,4,5])
print(array)

Expected output:

[1 2 3 4 5]

Creating Arrays with Ranges

NumPy makes it easy to create arrays with number sequences:

array = np.arange(5)
print(array)

Expected output:

[0 1 2 3 4]

Creating Special Arrays

Sometimes you need arrays filled with the same number:

array = np.ones(5)
print(array)

Expected output:

[1. 1. 1. 1. 1.]

:bulb: Arrays of ones and zeros are super useful when you need to initialize data or create placeholder arrays!

python

array = np.zeros(5)
print(array)

Expected output:

output

[0. 0. 0. 0. 0.]

Creating 2D Arrays (Tables)

You can create arrays that look like tables with rows and columns:

python

array2D = np.array([[1,2,3],[4,5,6]])
print(array2D)

Expected output:

output

[[1 2 3]
 [4 5 6]]

You can also create 2D arrays filled with ones or zeros:

python

array2D = np.zeros((2,3))
print(array2D)

Expected output:

output

[[0. 0. 0.]
 [0. 0. 0.]]

note For 2D arrays, (2,3) means 2 rows and 3 columns. Always remember: rows first, then columns!

:mag: NumPy Array Properties

Every NumPy array has special properties that tell us about its structure:

• dtype: What type of data is stored (integers, decimals, etc.)
• ndim: Number of dimensions (1D, 2D, 3D, etc.)
• shape: Size of each dimension
• size: Total number of elements

Let's explore these properties:

print(array.dtype)
print(array2D.dtype)

Expected output:

float64
float64

:information_source: Info float64 means the array stores decimal numbers with high precision

print(array.ndim)
print(array2D.ndim)

Expected output:

1
2

print(array.shape)
print(array2D.shape)

Expected output:

(5,)
(2, 3)

print(array.size)
print(array2D.size)

Expected output:

5
6

:emoji: Random Numbers in NumPy

NumPy includes tools for generating random numbers, which is useful for simulations and testing:

from numpy import random

Generate Random Decimal Numbers

Create a random decimal between 0 and 1:

print(random.rand())

Expected output:

0.3567409764700181

Generate Random Integers

Create a random whole number:

print(random.randint(10))

Expected output:

7

:bulb: random.randint(10) gives you a random number from 0 to 9 (not including 10)

:emoji: Access Items in NumPy Arrays

Accessing Single Items

Just like Python lists, you can get specific items using their position (index):

python

array = np.array([1,2,3,4,5])
print(array[1])

Expected output:

output

2

Accessing Items in 2D Arrays

For 2D arrays, specify both row and column:

python

array2D = np.array([[1,2,3],[4,5,6]])
print(array2D[1,2])

Expected output:

output

6

Accessing Multiple Items (Slicing)

You can get multiple items at once using slicing: note Remember the slicing rules:

• [start:end] - gets items from start to end-1
• [:end] - gets items from beginning to end-1
• [start:] - gets items from start to the end

print(array[1:3])
print(array[1:])
print(array[:3])

Expected output:

[2 3]
[2 3 4 5]
[1 2 3]

Slicing 2D Arrays

You can slice 2D arrays to get specific rows and columns:

print(array2D[:,1:3])

Expected output:

[[2 3]
 [5 6]]

:information_source: Info The : means "all rows", and 1:3 means "columns 1 and 2" (remember, 3 is not included!)

:emoji: Filter Items in NumPy Arrays

Filtering helps you find specific data that meets certain conditions. This is super useful in data analysis!

How Filtering Works

1. First, create a condition that returns True or False for each element
2. Then, use that condition to filter your data

Let's see how it works:

array > 2

Expected output:

array([False, False, True, True, True])

Now use this condition to get only the values greater than 2:

array[array > 2]

Expected output:

array([3, 4, 5])

Filtering 2D Arrays

The same technique works for 2D arrays:

array2D > 2

Expected output:

array([[False, False, True],
       [True, True, True]])

array2D[array2D > 2]

Expected output:

array([3, 4, 5, 6])

:memo: When you filter a 2D array, the result becomes a 1D array with all matching values!

:emoji: Mathematical Operations on NumPy Arrays

One of NumPy's superpowers is doing math with entire arrays at once! This is much faster than using loops.

Basic Math Operations

python

print(array + 10)

Expected output:

output

[11 12 13 14 15]

tip This adds 10 to EVERY element in the array - no loops needed!

print(array * 5)

Expected output:

[ 5 10 15 20 25]

Operations Between Arrays

You can also do math between two arrays:

print(array + array)

Expected output:

[ 2  4  6  8 10]

Saving Results

To save your calculation results, assign them back to a variable:

array = (array + array)
print(array)

Expected output:

[2 4 6 8 10]

:memo: Summary

In this lesson, we learned:

:white_check_mark: NumPy basics: A powerful library for working with numbers and arrays :white_check_mark: Creating arrays: Using np.array(), np.arange(), np.ones(), and np.zeros() :white_check_mark: Array properties: Understanding dtype, ndim, shape, and size :white_check_mark: Accessing data: Using indexing and slicing to get specific values :white_check_mark: Filtering: Finding data that meets certain conditions :white_check_mark: Math operations: Performing calculations on entire arrays at once

NumPy is your foundation for data science - master it, and you'll be ready for more advanced tools!

:movie_camera: Video

:emoji: AI Prompt

Code with AI: Try using AI to work with NumPy.

Prompts:

• "How do I create a NumPy array in Python?"
• "Generate Python code to calculate the mean, median, and mode of a NumPy array."
• "Show me how to filter a NumPy array to find values greater than 50"
• "Create a 3x3 NumPy array and multiply it by 2"

:dart: Practice Activities

1. Array Creation Challenge: Create three different arrays:

   • An array with numbers from 10 to 20
   • A 3x4 array filled with zeros
   • A 2D array with your favorite numbers

2. Filter Practice: Create an array of test scores and:

   • Find all scores above 80
   • Find all scores between 60 and 90
   • Count how many scores are passing (above 70)

3. Math Magic: Create two arrays and:

   • Add them together
   • Multiply one by 3
   • Calculate the average of all values

4. Real-World Application: Think of a situation where you'd need to:

   • Store data in a table format (2D array)
   • Filter that data based on conditions
   • Perform calculations on the entire dataset