Book Image

Python Data Science Essentials

Book Image

Python Data Science Essentials

Overview of this book

The book starts by introducing you to setting up your essential data science toolbox. Then it will guide you across all the data munging and preprocessing phases. This will be done in a manner that explains all the core data science activities related to loading data, transforming and fixing it for analysis, as well as exploring and processing it. Finally, it will complete the overview by presenting you with the main machine learning algorithms, the graph analysis technicalities, and all the visualization instruments that can make your life easier in presenting your results. In this walkthrough, structured as a data science project, you will always be accompanied by clear code and simplified examples to help you understand the underlying mechanics and real-world datasets.

Related Content you might be interested in

Current Title:

Small book image
Python Data Science Essentials
Apr, 2015 | 258 pages
No titles found
Table of Contents (13 chapters)
Python Data Science Essentials
Python Data Science Essentials
Credits
Credits
About the Authors
About the Authors
About the Reviewers
About the Reviewers
www.PacktPub.com
www.PacktPub.com
Preface
Preface
Free Chapter
1
First Steps
First Steps
Introducing data science and Python
Installing Python
Scientific distributions
Introducing IPython
Summary
2
Data Munging
Data Munging
The data science process
Data loading and preprocessing with pandas
Working with categorical and textual data
Data processing with NumPy
Creating NumPy arrays
NumPy fast operation and computations
Summary
3
The Data Science Pipeline
The Data Science Pipeline
Introducing EDA
Feature creation
Dimensionality reduction
The detection and treatment of outliers
Scoring functions
Testing and validating
Cross-validation
Hyper-parameters' optimization
Feature selection
Summary
4
Machine Learning
Machine Learning
Linear and logistic regression
Naive Bayes
The k-Nearest Neighbors
Advanced nonlinear algorithms
Ensemble strategies
A peek into Natural Language Processing (NLP)
An overview of unsupervised learning
Summary
5
Social Network Analysis
Social Network Analysis
Introduction to graph theory
Graph algorithms
Graph loading, dumping, and sampling
Summary
6
Visualization
Visualization
Introducing the basics of matplotlib
Selected graphical examples with pandas
Advanced data learning representation
Summary
Index
Index

Chapter 3. The Data Science Pipeline

Until now, we explored how to load data into Python and process it up to a point to create a dataset as a bidimensional NumPy array of numeric values. At this point, we are ready to get fully immersed into data science and extract meaning from data and potential data products. This chapter and the next chapter on machine learning are the most challenging sections of the entire book.

In this chapter, you will learn how to:

  • Briefly explore data and create new features

  • Reduce the dimensionality of data

  • Spot and treat outliers

  • Decide on the score or loss metrics that are the best for your project

  • Apply the scientific methodology and effectively test the performance of your machine learning hypothesis

  • Select the best feature set

  • Optimize your learning parameters

Previous Section
End of Section 1
Next Section