课程

Python 中的探索性数据分析

中级技能水平

更新时间 2026年4月

学习如何使用 Python 中的探索性数据分析（EDA）来探索、可视化并提取数据洞察。

免费开始课程

PythonExploratory Data Analysis

4小时

14 视频

49 道练习

4,150 XP

100K+

成就证明

深受数千家公司学习者的喜爱

需要团队培训？

企业版试用

课程描述

您已经拿到一份有意思的数据——应该如何开始分析？本课程将带您完成探索与分析数据的流程：从理解数据集中包含的内容，到将探索结果融入数据科学工作流。

以失业数据和机票价格为例，您将使用 Python 对数据进行汇总与验证，计算、识别并替换缺失值，同时清洗数值型与分类型数据。在整个课程中，您会使用 Seaborn 创建高质量可视化，帮助理解各个变量及其相互关系。

最后，课程将展示如何把探索结果纳入数据科学工作流，例如创建新特征、平衡分类特征，并基于发现提出可检验的假设。

完成本课程后，您将能够自信地在 Python 中开展探索性数据分析（EDA）。您也可以用可视化清晰地向他人解释您的发现，并为进一步获取洞见提出下一步建议！视频包含实时字幕，您可以点击视频左下角的 "Show transcript" 展开查看。课程术语表位于右侧的资源部分。如需获得 CPE 学分，您需要完成课程并在合格评估中取得 70% 的成绩。您可以点击右侧的 CPE 学分提示进入评估。

先决条件

Introduction to Statistics in Python Introduction to Data Visualization with Seaborn

1

Getting to Know a Dataset

What's the best way to approach a new dataset? Learn to validate and summarize categorical and numerical data and create Seaborn visualizations to communicate your findings.

Initial exploration

Functions for initial exploration

Counting categorical values

Global unemployment in 2021

Data validation

Detecting data types

Validating continents

Validating range

Data summarization

Summaries with .groupby() and .agg()

Named aggregations

Visualizing categorical summaries

2

Data Cleaning and Imputation

Exploring and analyzing data often means dealing with missing values, incorrect data types, and outliers. In this chapter, you’ll learn techniques to handle these issues and streamline your EDA processes!

Addressing missing data

Dealing with missing data

Strategies for remaining missing data

Imputing missing plane prices

Converting and analyzing categorical data

Finding the number of unique values

Flight duration categories

Adding duration categories

Working with numeric data

Flight duration

Adding descriptive statistics

Handling outliers

What to do with outliers

Identifying outliers

Removing outliers

3

Relationships in Data

Variables in datasets don't exist in a vacuum; they have relationships with each other. In this chapter, you'll look at relationships across numerical, categorical, and even DateTime data, exploring the direction and strength of these relationships as well as ways to visualize them.

Patterns over time

Importing DateTime data

Updating data type to DateTime

Visualizing relationships over time

Correlation

Interpreting a heatmap

Visualizing variable relationships

Visualizing multiple variable relationships

Factor relationships and distributions

Categorical data in scatter plots

Exploring with KDE plots

4

Turning Exploratory Analysis into Action

Exploratory data analysis is a crucial step in the data science workflow, but it isn't the end! Now it's time to learn techniques and considerations you can use to successfully move forward with your projects after you've finished exploring!

Considerations for categorical data

Checking for class imbalance

Cross-tabulation

Generating new features

Extracting features for correlation

Calculating salary percentiles

Categorizing salaries

Generating hypotheses

Comparing salaries

Choosing a hypothesis

Congratulations

Python 中的探索性数据分析

课程完成

获得成就证明

将此证书添加到您的 LinkedIn 档案、简历或履历中
在社交媒体和绩效评估中分享立即注册

加入超过19百万学习者，今天就开始Python 中的探索性数据分析！

通过 DataCamp for Mobile 提升您的数据技能

随时随地通过我们的移动课程和每日 5 分钟编程挑战提升技能。