Pandas

🐼 Pandas: Your Data’s Best Friend

The Story of the Magic Spreadsheet

Imagine you have a giant box of LEGO bricks scattered all over your room. Finding the red ones? Nightmare! Sorting by size? Hours of work! Now imagine a magic helper that can instantly find, sort, combine, and organize ALL your bricks in seconds.

That’s Pandas! 🎉

Pandas is like having a super-smart assistant for your data. It takes messy information and makes it neat, organized, and easy to understand.

🎯 What We’ll Learn

graph TD
    A["🐼 Pandas Basics"] --> B["📊 DataFrames"]
    B --> C["🔧 Data Manipulation"]
    C --> D["📦 GroupBy Magic"]
    D --> E["🔗 Merge & Join"]
    E --> F["📅 DateTime Handling"]

📚 Chapter 1: Meet Pandas

What is Pandas?

Think of Pandas as a super-powered spreadsheet that lives inside Python. Just like how Excel has rows and columns, Pandas has them too—but with superpowers!

Real Life Examples:

• Netflix uses it to organize viewer data
• Banks use it to track transactions
• Scientists use it to analyze experiments

Your First Pandas Code

import pandas as pd

# Create a simple table
data = {
    'Name': ['Anna', 'Bob', 'Cara'],
    'Age': [10, 12, 11],
    'Score': [95, 88, 92]
}

df = pd.DataFrame(data)
print(df)

Output:

   Name  Age  Score
0  Anna   10     95
1   Bob   12     88
2  Cara   11     92

🎈 Think of it: Each row is a person, each column is information about them.

📊 Chapter 2: The DataFrame - Your Data Table

What’s a DataFrame?

A DataFrame is like a table in a notebook. It has:

• Rows = Each item (like each student)
• Columns = Information about items (like name, age, score)
• Index = Row numbers (like seat numbers)

Creating DataFrames

Method 1: From a Dictionary

students = {
    'Name': ['Emma', 'Liam'],
    'Grade': ['A', 'B']
}
df = pd.DataFrame(students)

Method 2: From a List

data = [
    ['Emma', 'A'],
    ['Liam', 'B']
]
df = pd.DataFrame(data,
    columns=['Name', 'Grade'])

Selecting Data

# Get one column
df['Name']

# Get multiple columns
df[['Name', 'Grade']]

# Get one row by position
df.iloc[0]  # First row

# Get rows by condition
df[df['Grade'] == 'A']

🧙‍♂️ Magic Tip: iloc = position (like “item 0”), loc = label (like “row named X”)

🔧 Chapter 3: Data Manipulation

The Art of Shaping Data

Imagine you’re a chef preparing ingredients. Sometimes you need to:

• Add new ingredients (new columns)
• Remove bad parts (drop columns/rows)
• Change how things look (transform data)
• Filter out what you don’t need

Adding New Columns

df['Bonus'] = df['Score'] * 0.1

# Or with a condition
df['Pass'] = df['Score'] >= 60

Removing Data

# Drop a column
df = df.drop('Bonus', axis=1)

# Drop a row
df = df.drop(0, axis=0)

# Drop rows with missing values
df = df.dropna()

Changing Values

# Replace values
df['Grade'] = df['Grade'].replace(
    'F', 'Fail')

# Apply a function
df['Name'] = df['Name'].str.upper()

Filtering Data

# Students who passed
passed = df[df['Score'] >= 60]

# Multiple conditions
stars = df[
    (df['Score'] >= 90) &
    (df['Age'] <= 12)
]

🎯 Remember: & means AND, | means OR

📦 Chapter 4: GroupBy - Sorting into Buckets

The Bucket Story

Imagine you have a basket of mixed fruits and you want to:

1. Group them by type (apples together, oranges together)
2. Count how many of each
3. Find the biggest one in each group

That’s exactly what GroupBy does!

Basic GroupBy

# Sample data
sales = pd.DataFrame({
    'Store': ['A', 'B', 'A', 'B'],
    'Product': ['Apple', 'Apple',
                'Banana', 'Banana'],
    'Amount': [10, 15, 20, 25]
})

# Group by Store and sum
by_store = sales.groupby('Store')
print(by_store['Amount'].sum())

Output:

Store
A    30
B    40

Multiple Aggregations

# Get multiple stats at once
stats = sales.groupby('Store').agg({
    'Amount': ['sum', 'mean', 'count']
})

Group by Multiple Columns

detailed = sales.groupby(
    ['Store', 'Product']
)['Amount'].sum()

🪣 Think of it: GroupBy = Put similar things in buckets, then do math on each bucket!

🔗 Chapter 5: Merge and Join

The Puzzle Piece Story

Imagine you have two puzzle pieces that belong together:

• Piece 1: Student names and their IDs
• Piece 2: IDs and their test scores

To see “which student got which score,” you need to connect the pieces using the ID!

Types of Joins

graph TD
    A["Two Tables"] --> B["Inner Join"]
    A --> C["Left Join"]
    A --> D["Right Join"]
    A --> E["Outer Join"]
    B --> F["Only matching rows"]
    C --> G["All left + matching right"]
    D --> H["All right + matching left"]
    E --> I["All rows from both"]

Merge Example

# Table 1: Students
students = pd.DataFrame({
    'ID': [1, 2, 3],
    'Name': ['Anna', 'Bob', 'Cara']
})

# Table 2: Scores
scores = pd.DataFrame({
    'ID': [1, 2, 4],
    'Score': [95, 88, 75]
})

# Inner join (only matching IDs)
result = pd.merge(
    students, scores,
    on='ID', how='inner'
)

Result:

   ID  Name  Score
0   1  Anna     95
1   2   Bob     88

Different Join Types

# Left join - keep all students
left = pd.merge(
    students, scores,
    on='ID', how='left'
)

# Outer join - keep everyone
outer = pd.merge(
    students, scores,
    on='ID', how='outer'
)

🧩 Remember:

• inner = Only matches
• left = All from left table
• right = All from right table
• outer = Everything from both

📅 Chapter 6: DateTime Handling

Time is Data Too!

Dates and times are special. They’re not just numbers or text—they have meaning! Pandas understands this.

Examples of date questions:

• “How many sales in January?”
• “What day had the most visitors?”
• “How many hours between these events?”

Converting to DateTime

# Create dates from strings
df = pd.DataFrame({
    'date': ['2024-01-15',
             '2024-02-20',
             '2024-03-25']
})

# Convert to datetime
df['date'] = pd.to_datetime(df['date'])

Extracting Date Parts

# Get year, month, day
df['year'] = df['date'].dt.year
df['month'] = df['date'].dt.month
df['day'] = df['date'].dt.day

# Get day of week (0=Monday)
df['weekday'] = df['date'].dt.dayofweek

# Get day name
df['day_name'] = df['date'].dt.day_name()

Date Math

# Add days
df['next_week'] = df['date'] + \
    pd.Timedelta(days=7)

# Find difference
df['diff'] = df['date'].diff()

# Resample by month
monthly = df.set_index('date').\
    resample('M').sum()

Filtering by Date

# Sales in 2024
sales_2024 = df[
    df['date'].dt.year == 2024
]

# Between two dates
jan_sales = df[
    (df['date'] >= '2024-01-01') &
    (df['date'] <= '2024-01-31')
]

📆 Pro Tip: Always convert date strings to datetime FIRST, then do operations!

🎉 You Did It!

You’ve learned the six superpowers of Pandas:

🚀 Quick Reference

import pandas as pd

# Create DataFrame
df = pd.DataFrame(data)

# Select
df['column']      # One column
df[['a', 'b']]    # Multiple columns
df.iloc[0]        # Row by position
df.loc[0]         # Row by label

# Filter
df[df['col'] > 5]

# Group
df.groupby('col').sum()

# Merge
pd.merge(df1, df2, on='key')

# DateTime
pd.to_datetime(df['date'])
df['date'].dt.year

Remember: Data is like LEGO bricks. Pandas helps you build anything you imagine! 🧱✨