
Data Science from Scratch: First Principles with Python, 2nd Edition
Data Science from Scratch: First Principles with Python is a second-edition guide to the maths, statistics and programming ideas behind data science. It describes implementing algorithms from scratch and covers topics from data preparation and machine learning to natural language processing and network analysis.
At a glance
- Updated for Python 3.6, according to the listing.
- Explains data science concepts by implementing tools and algorithms from scratch.
- Topics include linear algebra, statistics, probability, data preparation and machine learning.
- Further coverage includes recommender systems, natural language processing, network analysis, MapReduce and databases.
Often bought together
Pros and cons
Pros
- Combines programming with foundational maths and statistics rather than focusing only on libraries and tools.
- The described algorithms span classical machine learning methods, neural networks and clustering.
- Covers practical data handling alongside topics such as databases and network analysis.
Technical specifications
9 specifications| Edition | 2nd |
|---|---|
| Publication date | 30 April 2019 |
| Publisher | O'Reilly Media, Inc, USA |
| Language | English |
| Print length | 403 pages |
| ISBN-10 | 1492041130 |
| ISBN-13 | 978-1492041139 |
| Weight listed | 712 g |
|---|---|
| Dimensions | 17.53 x 2.29 x 23.11 cm |
What to know before buying
- The listing frames the book for readers with some programming skills and an aptitude for mathematics. Its Python update is specified as Python 3.6, so buyers seeking guidance for a newer Python version should check whether the material suits their needs.
Who is it for?
A good fit for
- Readers with some programming experience who want to understand data science algorithms from first principles.
- Learners looking for a broad introduction to data preparation, statistics and machine learning in Python.
- Readers interested in a single book that also touches on NLP, recommender systems and network analysis.
Less suited to
- Readers seeking a no-prerequisites introduction, given the stated expectation of some programming skills and mathematical aptitude.
- Buyers who need a guide explicitly updated for Python versions newer than Python 3.6.
Customer reviews on Amazon
- 5 stars
71% 71% - 4 stars
16% 16% - 3 stars
8% 8% - 2 stars
2% 2% - 1 star
3% 3%
Aggregate customer ratings are largely in the five-star category, with a smaller spread across lower ratings. This offers a broadly favourable but not universal signal. The ratings do not explain what readers liked or disliked, so they cannot establish how well the book's Python 3.6 examples or broad topic coverage will suit a particular learner.
Frequently asked questions
What edition is listed?
The listing specifies the 2nd edition, published on 30 April 2019.
Which Python version does the listing mention?
It says the second edition is updated for Python 3.6.
What topics does the book cover?
The supplied description lists maths and statistics foundations, data handling, machine learning, recommender systems, natural language processing, network analysis, MapReduce and databases.
Is this presented as a beginner book with no prerequisites?
No. The description says readers should have some programming skills and an aptitude for mathematics.
Alternative options

Machine Learning with PyTorch and Scikit-Learn: Develop machine learning and deep learning models with Python
$75.99

Step-by-Step Electrical Engineering: Learn and Apply Core Concepts Quickly. Build Your Knowledge with Practical Examples - No Background Required! Perfect for Aspiring Engineers and Enthusiasts!
$26.99

Generative AI with LangChain - Second Edition: Build production-ready LLM applications and advanced agents using Python, LangChain, and LangGraph
$85.68

Ultimate Data Engineering with Databricks: Develop Scalable Data Pipelines Using Data Engineering's Core Tenets Such as Delta Tables, Ingestion, ... Security, and Scalability (English Edition)
$63.79

Ultimate Azure Data Engineering: Build Robust Data Engineering Systems on Azure with SQL, ETL, Data Modeling, and Power BI for Business Insights and Crack Azure Certifications (English Edition)
$59.42

Data Analysis in Microsoft Excel: Deliver Awesome Analytics in 3 Easy Steps Using VLOOKUPS, Pivot Tables, Charts And More
$32.99

Prisoners of Geography: 10-Year Anniversary Edition: Ten maps that tell you everything you need to know about global politics
$17.33

Building AI Agents with LLMs, RAG, and Knowledge Graphs: A practical guide to autonomous and modern AI agents
$87.56

























