Introduction to Pandas and Working with Tabular Data

Why Pandas?

When working with data, you’ll almost always deal with tables — datasets with rows and columns, like spreadsheets. Enter Pandas: Python’s go-to library for handling structured (tabular) data. It lets you clean, filter, reshape, group, and analyze data with minimal code.

Whether you’re loading a CSV, removing null values, or summarizing trends, Pandas is your best friend.

Getting Started with Pandas

To start using Pandas:

pythonCopyEditimport pandas as pd

You’ll mostly work with DataFrames (tables) and Series (single columns).

pythonCopyEditdata = {
    'Name': ['Anay', 'Sara', 'John'],
    'Age': [12, 25, 22]
}

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

Output:

nginxCopyEdit   Name  Age
0  Anay   12
1  Sara   25
2  John   22

Reading and Writing Data

Pandas can handle many file formats, but CSV is the most common.

Load a CSV:

pythonCopyEditdf = pd.read_csv('students.csv')

Save to a CSV:

pythonCopyEditdf.to_csv('output.csv', index=False)

You can also load Excel, JSON, SQL, etc.

Exploring the Data

Once you load a dataset, your first job is to understand it.

pythonCopyEditdf.head()        # Shows first 5 rows  
df.tail()        # Last 5 rows  
df.shape         # (rows, columns)  
df.info()        # Summary of data types and nulls  
df.describe()    # Stats for numerical columns

These quick commands give you a strong overview.

Selecting Columns and Rows

Access a single column (Series):

pythonCopyEditdf['Age']            # Using bracket notation
df.Age               # Dot notation (only for simple names)

Access multiple columns:

pythonCopyEditdf[['Name', 'Age']]

Filter rows:

pythonCopyEditdf[df['Age'] > 18]   # Students older than 18

Cleaning Data (Common Tasks)

Real-world data is messy. Here’s how Pandas helps:

• Missing values:

pythonCopyEditdf.isnull().sum()            # Count nulls
df.dropna()                  # Remove rows with nulls
df.fillna(0)                 # Replace nulls with 0

• Renaming columns:

pythonCopyEditdf.rename(columns={'Age': 'StudentAge'}, inplace=True)

• Changing types:

pythonCopyEditdf['StudentAge'] = df['StudentAge'].astype(int)

• Dropping duplicates:

pythonCopyEditdf.drop_duplicates(inplace=True)

Sorting and Grouping

Sort rows:

pythonCopyEditdf.sort_values(by='Age', ascending=False)

Group and summarize:

pythonCopyEditdf.groupby('Age').count()         # Count rows by age
df.groupby('Age').mean()          # Average by age

This is how you can quickly spot trends and patterns.

Real-World Use Case Example

Imagine you’re analyzing online orders. With Pandas, you can:

• Load the CSV
• Clean missing values
• Filter for orders above ₹1,000
• Group by product category and find average spend
• Sort to find the top-performing categories

All with just a few lines of code.

Quick Summary

Final Thoughts

Pandas makes data exploration intuitive and fast. It’s designed to minimize friction between your brain and your data. Once you master it, tasks that once felt overwhelming — like wrangling giant datasets — will become second nature.

In the next article, we’ll focus on data visualization — turning your tables into charts that tell powerful stories.

Next Up: Data Visualization with Matplotlib and Seaborn