🔍 Pandas Foundations: Inspecting and Exporting Your Data

The Detective’s Toolkit 🕵️

Imagine you just received a mystery box filled with data. Before you can solve any puzzle, you need to look inside and understand what you have. That’s exactly what Pandas helps you do!

Think of a DataFrame like a filing cabinet with labeled drawers. Before organizing anything, you need to:

• See what types of things are in each drawer
• Get a quick summary of everything
• Check if any drawers are empty
• Eventually, share your organized files with others

Let’s open our detective toolkit!

📊 What Type Is This? The dtypes Attribute

The Label Maker

Every column in your DataFrame has a type — like labels on jars in your kitchen.

• Numbers (int64, float64) → For math stuff
• Text (object) → Words and sentences
• Dates (datetime64) → Calendar things
• True/False (bool) → Yes or No answers

See Your Types

import pandas as pd

# Our toy box of data
df = pd.DataFrame({
    'name': ['Anna', 'Bob', 'Cara'],
    'age': [10, 12, 11],
    'height': [4.2, 4.8, 4.5],
    'has_pet': [True, False, True]
})

# Look at all the labels!
print(df.dtypes)

Output:

name       object
age         int64
height    float64
has_pet      bool
dtype: object

🎯 Quick Translation

📋 The Full Report: info() Method

Your Data’s ID Card

The info() method is like asking your data: “Tell me everything about yourself!”

It shows:

• ✅ How many rows and columns
• ✅ Column names
• ✅ Data types
• ✅ Missing values count
• ✅ Memory usage

df.info()

Output:

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 3 entries, 0 to 2
Data columns (total 4 columns):
 #   Column   Non-Null Count  Dtype
---  ------   --------------  -----
 0   name     3 non-null      object
 1   age      3 non-null      int64
 2   height   3 non-null      float64
 3   has_pet  3 non-null      bool
dtypes: bool(1), float64(1), int64(1), object(1)
memory usage: 227.0+ bytes

🔑 Key Things to Notice

• 3 entries = 3 rows of data
• Non-Null Count = How many cells have real values
• memory usage = How much computer space it takes

📈 Quick Math Summary: describe() Method

The Statistics Wizard

Want quick math facts about your numbers? describe() does the magic!

print(df.describe())

Output:

            age    height
count  3.000000  3.000000
mean  11.000000  4.500000
std    1.000000  0.300000
min   10.000000  4.200000
25%   10.500000  4.350000
50%   11.000000  4.500000
75%   11.500000  4.650000
max   12.000000  4.800000

🎯 What Each Row Means

graph LR
    A[describe output] --> B[count: How many values]
    A --> C[mean: The average]
    A --> D[std: How spread out]
    A --> E[min: Smallest value]
    A --> F[25%: Quarter mark]
    A --> G[50%: Middle value]
    A --> H[75%: Three-quarter mark]
    A --> I[max: Biggest value]

Include Text Columns Too!

# See ALL columns, not just numbers
print(df.describe(include='all'))

This adds unique, top, and freq for text columns!

🆕 Creating Empty DataFrames

Starting Fresh

Sometimes you need a blank canvas before painting. Here’s how to create an empty DataFrame:

# Method 1: Completely empty
empty_df = pd.DataFrame()

# Method 2: Empty with column names
empty_df = pd.DataFrame(
    columns=['name', 'score', 'grade']
)

# Method 3: Empty with specific types
empty_df = pd.DataFrame({
    'name': pd.Series(dtype='object'),
    'score': pd.Series(dtype='int64')
})

Why Create Empty DataFrames?

• 📦 To fill with data later
• 🔄 As a starting point for loops
• 📝 To define your structure first

❓ Is This DataFrame Empty?

The Empty Check

Before working with data, it’s smart to check if there’s actually anything there!

# Create an empty DataFrame
empty_df = pd.DataFrame()

# Check if it's empty
if empty_df.empty:
    print("Nothing here! 📭")
else:
    print("We have data! 📬")

Output: Nothing here! 📭

Two Ways to Check

# Method 1: The .empty attribute
df.empty  # Returns True or False

# Method 2: Check the length
len(df) == 0  # Also True or False

Real-World Example

# After filtering, check if results exist
results = df[df['age'] > 100]

if results.empty:
    print("No one is over 100 years old!")
else:
    print(f"Found {len(results)} centenarians!")

📤 Exporting Your Data

Sharing Is Caring

Once you’ve cleaned and organized your data, you want to save it or share it. Pandas can export to many formats!

Export to CSV (Most Common)

# Save to a CSV file
df.to_csv('my_data.csv', index=False)

# index=False means don't save row numbers

Export to Excel

# Save to Excel file
df.to_excel('my_data.xlsx', index=False)

Export to JSON

# Save as JSON
df.to_json('my_data.json')

# Pretty format for reading
df.to_json('my_data.json', indent=2)

More Export Options

graph LR
    A[DataFrame] --> B[.to_csv - Spreadsheets]
    A --> C[.to_excel - Microsoft Excel]
    A --> D[.to_json - Web apps]
    A --> E[.to_html - Web pages]
    A --> F[.to_sql - Databases]
    A --> G[.to_pickle - Python storage]

Quick Export Reference

🎯 Quick Summary Flow

graph TD
    A[Get DataFrame] --> B[Check dtypes]
    B --> C[Run info]
    C --> D[Run describe]
    D --> E{Is it empty?}
    E -->|Yes| F[Add data first!]
    E -->|No| G[Work with data]
    G --> H[Export when done]
    H --> I[CSV/Excel/JSON]

🌟 You Did It!

You now know how to:

✅ See data types with .dtypes

✅ Get full info with .info()

✅ Get statistics with .describe()

✅ Create empty DataFrames

✅ Check if DataFrame is empty with .empty

✅ Export data to CSV, Excel, JSON, and more!

You’re ready to inspect and export like a pro! 🚀