NumPy, Pandas, and Scikit-learn Masterclass

Applied data wrangling and machine learning projects with Python (English Edition)

Gary Hutson

• Publisher: BPB Publications
• Year: 2025
• eBook ISBN: 9789365898293
• Subtopic: Data Processing

Description
Data is the driving force of today's digital economy, and the ability to wrangle, analyze, and model it effectively has become a vital skill across industries. Python's powerful ecosystem, led by libraries like NumPy, Pandas, and Scikit-learn, enables professionals to transform raw datasets into meaningful insights and build ML solutions that solve real-world problems.

This book offers a practical, hands-on journey into mastering these libraries step-by-step. You will begin with NumPy and Pandas, learning how to manipulate arrays, DataFrames, time series, and large datasets efficiently. The focus then shifts to Scikit-learn, where you will explore classification, regression, clustering, dimensionality reduction, and time series modeling. Along the way, you will explore the practical case studies, including thyroid disease prediction, customer segmentation, and housing price estimation. You will also explore topics such as hyperparameter tuning, ensemble methods, pipelines, and deep neural networks, followed by guidance on deploying ML models with Flask, FastAPI, Docker, and Swagger.

By the end of this book, you will be confident in applying the core data science libraries of Python to real-world problems. You will be able to clean and transform complex datasets, build and optimize robust ML models, and deploy them into production environments as scalable APIs. This book equips you with the practical skills needed to excel in data-driven roles and deliver impactful ML solutions.

What you will learn
? Manipulate arrays and datasets using NumPy and Pandas effectively.
? Preprocess data and build models with Scikit-learn workflows.
? Apply regression, classification, dimensionality reduction, time series forecasting, deep learning, and clustering to real datasets.
? Handle missing values, time series, and large-scale data.
? Optimize performance with hyperparameter tuning and ensemble methods.
? Deploy ML models as scalable RESTful APIs.

Who this book is for
This book is for Python developers, data analysts, system administrators, cloud engineers, aspiring data scientists, and anyone looking to master data wrangling and ML. It is ideal for professionals seeking to transition into data-driven roles and apply practical ML solutions in their jobs.

Table of Contents
1. Overview of NumPy and Pandas
2. Introduction to Scikit-learn for Machine Learning
3. Supervised Binary Classification
4. Supervised Multi-class Classification
5. Customer Segmentation with Unsupervised Methods
6. House Price Estimation with Regression Methods
7. Handwritten Digits Dimensionality Reduction
8. Time Series with Scikit-learn
9. Model Improvement Strategies
10. Building Multi-step Pipelines
11. Getting Deep with Deep Neural Networks
12. Deploying Your Machine Learning Application
13. Machine Learning Future Trends and Ethical Considerations

Table of contents

• Cover Page
• Title Page
• Copyright Page
• Dedication
• About the Author
• About the Reviewers
• Acknowledgement
• Preface
• Table of Contents
• 1. Overview of NumPy and Pandas
• 2. Introduction to Scikit-learn for Machine Learning
• 3. Supervised Binary Classification
• 4. Supervised Multi-class Classification
• 5. Customer Segmentation with Unsupervised Methods
• 6. House Price Estimation with Regression Methods
• 7. Handwritten Digits Dimensionality Reduction
• 8. Time Series with Scikit-learn
• 9. Model Improvement Strategies
• 10. Building Multi-step Pipelines
• 11. Getting Deep with Deep Neural Networks
• 12. Deploying Your Machine Learning Application
• 13. Machine Learning Future Trends and Ethical Considerations
• Index