Master Pandas Basics: From DataFrames to Quick Data Insights

01 Important Preface

This is the first article of the Python data analysis practical series, focusing on a simple encounter with Pandas. If you are already proficient with Pandas, you can skim or skip this content.

Many beginners quickly learn Python syntax but then dive straight into the classic "Python for Data Analysis" book, finishing it without truly understanding how to apply the knowledge. The main obstacles are insufficient understanding and lack of practice, often leading to a "three examples, one confusion" trap.

To avoid this, beginners should not overwhelm themselves with multiple methods for a single problem before mastering any.

02 Pandas Introduction

Pandas is a professional data analysis tool built on NumPy, offering flexible and efficient handling of various datasets. It provides two primary data structures: DataFrame (similar to an Excel sheet) and Series (a column in that sheet).

Before processing data, it is crucial to plan the analysis, clarify its purpose, and outline the workflow.

03 Create, Read, and Store

1. Create

First, import the library: import pandas as pd Construct a DataFrame using a dictionary of lists:

If no index is specified, Pandas generates a default integer index starting from 0.

2. Read

Read CSV files (use engine='python' to avoid encoding issues):

Read Excel files similarly:

Both functions support additional parameters such as header, sep, and names.

3. Store

Saving data is straightforward:

04 Quick Data Overview

1. View head and tail

Use df.head() to see the first few rows and df.tail() for the last rows. Both accept an integer argument to specify the number of rows.

2. Inspect data types df.info() displays column data types, non‑null counts, and memory usage.

3. Summary statistics df.describe() provides key statistics for numeric columns.

05 Basic Column Operations

Think of column operations as the four SQL actions: add, delete, select, update .

1. Add

Assign a new column:

df['new_column'] = values

2. Delete

Use df.drop('column_name', axis=1, inplace=True) to remove a column.

3. Select

Single column: df['column']. Multiple columns:

df[['col1', 'col2']]

4. Update

Modify a column directly:

df['existing_column'] = new_values

06 Common Data Types and Operations

String

String operations are similar to native Python strings but require the .str accessor.

Example: removing a leading dash from a column of strings.

Numeric

Perform arithmetic directly on columns, e.g., adding a constant or computing a new metric:

df['sales'] = df['visitors'] * df['conversion_rate'] * df['avg_price']

Be careful with mixed types; convert percentage strings to floats before calculations.

Datetime

Convert string dates to datetime objects with pd.to_datetime() and perform date arithmetic.

Recap

Understand what Pandas is.

Learn how to create, read, and store data.

Quickly inspect data with head, tail, info, and describe.

Perform basic add, delete, select, and update operations on columns.

Familiarize with common data types: strings, numerics, and datetimes.

Reply "初识pandas" to obtain the full case dataset.