Book Image

Data Science Projects with Python

By : Stephen Klosterman
Book Image

Data Science Projects with Python

By: Stephen Klosterman

Overview of this book

Data Science Projects with Python is designed to give you practical guidance on industry-standard data analysis and machine learning tools, by applying them to realistic data problems. You will learn how to use pandas and Matplotlib to critically examine datasets with summary statistics and graphs, and extract the insights you seek to derive. You will build your knowledge as you prepare data using the scikit-learn package and feed it to machine learning algorithms such as regularized logistic regression and random forest. You’ll discover how to tune algorithms to provide the most accurate predictions on new and unseen data. As you progress, you’ll gain insights into the working and output of these algorithms, building your understanding of both the predictive capabilities of the models and why they make these predictions. By then end of this book, you will have the necessary skills to confidently use machine learning algorithms to perform detailed data analysis and extract meaningful insights from unstructured data.

Related Content you might be interested in

Current Title:

Small book image
Data Science Projects with Python
Stephen Klosterman
Apr, 2019 | 374 pages
Small book image
Hands-On Predictive Analytics with Python
Alvaro Fuentes
Dec, 2018 | 330 pages
Small book image
Applied Supervised Learning with Python
Benjamin Johnston | Ishita Mathur
Apr, 2019 | 404 pages
Small book image
Data Cleaning and Exploration with Machine Learning
Michael Walker
Aug, 2022 | 542 pages
Table of Contents (9 chapters)
Data Science Projects with Python
Data Science Projects with Python
Preface
Preface
Free Chapter
1
Data Exploration and Cleaning
Data Exploration and Cleaning
Introduction
Python and the Anaconda Package Management System
Different Types of Data Science Problems
Loading the Case Study Data with Jupyter and pandas
Data Quality Assurance and Exploration
Exploring the Financial History Features in the Dataset
Summary
2
Introduction toScikit-Learn and Model Evaluation
Introduction toScikit-Learn and Model Evaluation
Introduction
Exploring the Response Variable and Concluding the Initial Exploration
Introduction to Scikit-Learn
Model Performance Metrics for Binary Classification
Summary
3
Details of Logistic Regression and Feature Exploration
Details of Logistic Regression and Feature Exploration
Introduction
Examining the Relationships between Features and the Response
Univariate Feature Selection: What It Does and Doesn't Do
Summary
4
The Bias-Variance Trade-off
The Bias-Variance Trade-off
Introduction
Estimating the Coefficients and Intercepts of Logistic Regression
Cross Validation: Choosing the Regularization Parameter and Other Hyperparameters
Summary
5
Decision Trees and Random Forests
Decision Trees and Random Forests
Introduction
Decision trees
Random Forests: Ensembles of Decision Trees
Summary
6
Imputation of Missing Data, Financial Analysis, and Delivery to Client
Imputation of Missing Data, Financial Analysis, and Delivery to Client
Introduction
Review of Modeling Results
Dealing with Missing Data: Imputation Strategies
Final Thoughts on Delivering the Predictive Model to the Client
Summary
Appendix
Appendix
Chapter 1: Data Exploration and Cleaning
Chapter 2: Introduction to Scikit-Learn and Model Evaluation
Chapter 3: Details of Logistic Regression and Feature Exploration
Chapter 4: The Bias-Variance Trade-off
Chapter 5: Decision Trees and Random Forests
Chapter 6: Imputation of Missing Data, Financial Analysis, and Delivery to Client

Model Performance Metrics for Binary Classification

Before we start building predictive models in earnest, we would like to know how we can determine, once we've created a model, whether it is "good" in some sense of the word. As you may imagine, this question has received a lot of attention from researchers and practitioners. Consequently, there is a wide variety of model performance metrics to choose from.

Note

For an idea of the range of options, have a look at the scikit-learn model evaluation page: https://scikit-learn.org/stable/modules/model_evaluation.html#model-evaluation.

When selecting a model performance metric to assess the predictive quality of a model, it's important to keep two things in mind.

Appropriateness of the metric for the problem

Metrics are typically only defined for a specific class of problems, such as classification or regression. For a binary classification problem, several metrics characterize the correctness of the yes or no question that the model answers. An...

Previous Section
End of Section 5
Next Section