[Week 3] First Steps toward Machine Learning: Python Basics and Data Representation 🐍

Hello! Welcome to Week 3.

Over the past 2 weeks, we have seen the remarkable possibilities of Machine Learning.

For the next 4 weeks, we will learn the most important tool for building these models ourselves,

We will learn the Python programming language.

In this first session, we will explore why Python is essential in Machine Learning,

We will learn how to represent and store data in Python.

🌐 Module 1: Why Use Python for Machine Learning?

• Simple, intuitive syntax:
  • Easier to learn than other languages, with concise code
  • Lets you focus on implementing complex mathematical theory
• Powerful and Extensive Libraries:
  • NumPy: An essential library for scientific computing and multidimensional arrays
  • Pandas: A library optimized for tabular data
  • Matplotlib: A data visualization library
  • Scikit-Learn: A library of classical Machine Learning models
  • TensorFlow, PyTorch: Major libraries for Deep Learning models
• A Strong Community:
  • Many developers and researchers worldwide use Python.
  • Resources and help for solving problems are easy to find.

📦 Module 2: Variables, the Smallest Units of Data

• What is a variable?: A box with a name tag for storing data (values)
• Variables in Machine Learning:
  • A value occupying one cell in a dataset
  • For example, one student's 'study hours' or 'exam score'
• Data Types: The forms of data stored in variables
  • int: Integer (e.g., 92, -5, 0)
  • float: Floating-point number (a number with a decimal point, e.g., 10.5, 3.14)
  • str: String (characters enclosed in quotes, e.g., "Kim", "Pass")
  • bool: Boolean (only true or false, e.g., True, False)

# Store 'one student's' data in variables.

# 1. Store the floating-point (float) value 10.5 in a variable labeled 'study hours'
study_hours = 10.5

# 2. Store the integer (int) value 92 in a variable labeled 'exam score'
exam_score = 92

# Check the stored values
print("공부 시간:", study_hours)
print("시험 점수:", exam_score)

# Check variable data types using type()
print("\n'study_hours' 변수의 자료형:", type(study_hours))
print("'exam_score' 변수의 자료형:", type(exam_score))

공부 시간: 10.5
시험 점수: 92

'study_hours' 변수의 자료형: <class 'float'>
'exam_score' 변수의 자료형: <class 'int'>

# Store additional student information

# Store the string (str) value "Kim" in a variable labeled 'student name'
student_name = "Kim"

# Store the Boolean (bool) value True in a variable labeled 'pass status'
is_passed = True

# Check the stored values
print("학생 이름:", student_name)
print("합격 여부:", is_passed)

# Check data types
print("\n'student_name' 변수의 자료형:", type(student_name))
print("'is_passed' 변수의 자료형:", type(is_passed))

학생 이름: Kim
합격 여부: True

'student_name' 변수의 자료형: <class 'str'>
'is_passed' 변수의 자료형: <class 'bool'>

# Variables can also be used in calculations.
# For example, suppose this student answered a 5-point bonus question correctly.

print("기존 점수:", exam_score)

# Add 5 to the existing score and update it
final_score = exam_score + 5

print("최종 점수:", final_score)

기존 점수: 92
최종 점수: 97

🧺 Module 3: Data Structures, Containers for Datasets

How can we handle multiple students' data at once?

We need a larger container, a data structure, that can hold multiple values in one variable.

1. Lists

• Definition:
  • A data structure that stores multiple values in order
  • Delimited by square brackets []
• Lists in Machine Learning: Useful for grouping and processing data in order
• Example: Lists of all students' 'study hours' and 'exam scores'

# Create lists of 'study hours' and 'exam scores' for multiple students.
study_hours_list = [2.0, 4.5, 6.0, 8.5, 10.0]
exam_scores_list = [65, 78, 85, 95, 98]

print("공부 시간 리스트:", study_hours_list)
print("시험 점수 리스트:", exam_scores_list)

# Lists have an order, so specific values can be accessed by index.
# Python starts counting at 0!
print("\n첫 번째 학생의 공부 시간:", study_hours_list[0])
print("세 번째 학생의 시험 점수:", exam_scores_list[2])

# Use len() to check how many data items a list contains.
print("\n총 학생 수:", len(study_hours_list))

공부 시간 리스트: [2.0, 4.5, 6.0, 8.5, 10.0]
시험 점수 리스트: [65, 78, 85, 95, 98]

첫 번째 학생의 공부 시간: 2.0
세 번째 학생의 시험 점수: 85

총 학생 수: 5

# Suppose a new student's data have been added.
# Use .append() to add a value to the end of a list.

print("추가 전 공부 시간 리스트:", study_hours_list)

study_hours_list.append(12.0)
exam_scores_list.append(100)

print("추가 후 공부 시간 리스트:", study_hours_list)
print("추가 후 총 학생 수:", len(study_hours_list))

추가 전 공부 시간 리스트: [2.0, 4.5, 6.0, 8.5, 10.0]
추가 후 공부 시간 리스트: [2.0, 4.5, 6.0, 8.5, 10.0, 12.0]
추가 후 총 학생 수: 6

2. Dictionaries

• Definition:
  • A data structure storing 'Key'–'Value' pairs
  • Delimited by braces {}
  • Unordered
• Dictionaries in Machine Learning: Useful for grouping and processing data according to their relationships
• Example: A bundle of information containing one student's 'name', 'study hours', 'exam score', and 'pass status'

# Create a dictionary containing all information about one student.
student_A_info = {
    'name': 'Park',
    'study_hours': 8.5,
    'exam_score': 95,
    'is_passed': True
}

print("학생 A의 정보:", student_A_info)

# A dictionary accesses values using Keys rather than positions (indices).
print("\n학생 A의 이름:", student_A_info['name'])
print("학생 A의 시험 점수:", student_A_info['exam_score'])

학생 A의 정보: {'name': 'Park', 'study_hours': 8.5, 'exam_score': 95, 'is_passed': True}

학생 A의 이름: Park
학생 A의 시험 점수: 95

Final form: Combining lists and dictionaries

• Most data used in Machine Learning are represented as a 'list of dictionaries' or a 'dictionary of lists'.
• In particular, grouping each student's information (a dictionary) into one list is very common.
  • It resembles an Excel table.

# Group multiple students' information (dictionaries) in one list to represent the full dataset
full_dataset = [
    { 'name': 'Kim', 'study_hours': 2.0, 'exam_score': 65, 'is_passed': True },
    { 'name': 'Lee', 'study_hours': 4.5, 'exam_score': 78, 'is_passed': True },
    { 'name': 'Park', 'study_hours': 8.5, 'exam_score': 95, 'is_passed': True },
    { 'name': 'Choi', 'study_hours': 1.5, 'exam_score': 55, 'is_passed': False }
]

# Access the first student's ('Kim') information
first_student = full_dataset[0]
print("첫 번째 학생의 정보:", first_student)

# Access only the first student's 'exam score'
first_student_score = full_dataset[0]['exam_score']
print("첫 번째 학생의 시험 점수:", first_student_score)

첫 번째 학생의 정보: {'name': 'Kim', 'study_hours': 2.0, 'exam_score': 65, 'is_passed': True}
첫 번째 학생의 시험 점수: 65