Applied Supervised Learning with Python电子书

Preface

About the Book

About the Authors

Objectives

Audience

Approach

Hardware Requirements

Software Requirements

Conventions

Installation and Setup

Installing the Code Bundle

Additional Resources

Chapter 1

Python Machine Learning Toolkit

Introduction

Supervised Machine Learning

When to Use Supervised Learning

Why Python?

Jupyter Notebooks

Exercise 1: Launching a Jupyter Notebook

Exercise 2: Hello World

Exercise 3: Order of Execution in a Jupyter Notebook

Exercise 4: Advantages of Jupyter Notebooks

Python Packages and Modules

pandas

Loading Data in pandas

Exercise 5: Loading and Summarizing the Titanic Dataset

Exercise 6: Indexing and Selecting Data

Exercise 7: Advanced Indexing and Selection

pandas Methods

Exercise 8: Splitting, Applying, and Combining Data Sources

Lambda Functions

Exercise 9: Lambda Functions

Data Quality Considerations

Managing Missing Data

Class Imbalance

Low Sample Size

Activity 1: pandas Functions

Summary

Chapter 2

Exploratory Data Analysis and Visualization

Introduction

Exploratory Data Analysis (EDA)

Exercise 10: Importing Libraries for Data Exploration

Summary Statistics and Central Values

Standard Deviation

Percentiles

Exercise 11: Summary Statistics of Our Dataset

Missing Values

Finding Missing Values

Exercise 12: Visualizing Missing Values

Imputation Strategies for Missing Values

Exercise 13: Imputation Using pandas

Exercise 14: Imputation Using scikit-learn

Exercise 15: Imputation Using Inferred Values

Activity 2: Summary Statistics and Missing Values

Distribution of Values

Target Variable

Exercise 16: Plotting a Bar Chart

Categorical Data

Exercise 17: Datatypes for Categorical Variables

Exercise 18: Calculating Category Value Counts

Exercise 19: Plotting a Pie Chart

Continuous Data

Exercise 20: Plotting a Histogram

Exercise 21: Skew and Kurtosis

Activity 3: Visually Representing the Distribution of Values

Relationships within the Data

Relationship between Two Continuous Variables

Exercise 22: Plotting a Scatter Plot

Exercise 23: Correlation Heatmap

Exercise 24: Pairplot

Relationship between a Continuous and a Categorical Variable

Exercise 25: Bar Chart

Exercise 26: Box Plot

Relationship between Two Categorical Variables

Exercise 27: Stacked Bar Chart

Activity 4: Relationships Within the Data

Summary

Chapter 3

Regression Analysis

Introduction

Regression and Classification Problems

Data, Models, Training, and Evaluation

Linear Regression

Exercise 28: Plotting Data with a Moving Average

Activity 5: Plotting Data with a Moving Average

Least Squares Method

The scikit-learn Model API

Exercise 29: Fitting a Linear Model Using the Least Squares Method

Activity 6: Linear Regression Using the Least Squares Method

Linear Regression with Dummy Variables

Exercise 30: Introducing Dummy Variables

Activity 7: Dummy Variables

Parabolic Model with Linear Regression

Exercise 31: Parabolic Models with Linear Regression

Activity 8: Other Model Types with Linear Regression

Generic Model Training

Gradient Descent

Exercise 32: Linear Regression with Gradient Descent

Exercise 33: Optimizing Gradient Descent

Activity 9: Gradient Descent

Multiple Linear Regression

Exercise 34: Multiple Linear Regression

Autoregression Models

Exercise 35: Creating an Autoregression Model

Activity 10: Autoregressors

Summary

Chapter 4

Classification

Introduction

Linear Regression as a Classifier

Exercise 36: Linear Regression as a Classifier

Logistic Regression

Exercise 37: Logistic Regression as a Classifier – Two-Class Classifier

Exercise 38: Logistic Regression – Multiclass Classifier

Activity 11: Linear Regression Classifier – Two-Class Classifier

Activity 12: Iris Classification Using Logistic Regression

Classification Using K-Nearest Neighbors

Exercise 39: K-NN Classification

Exercise 40: Visualizing K-NN Boundaries

Activity 13: K-NN Multiclass Classifier

Classification Using Decision Trees

Exercise 41: ID3 Classification

Exercise 42: Iris Classification Using a CART Decision Tree

Summary

Chapter 5

Ensemble Modeling

Introduction

Exercise 43: Importing Modules and Preparing the Dataset

Overfitting and Underfitting

Underfitting

Overfitting

Overcoming the Problem of Underfitting and Overfitting

Bagging

Bootstrapping

Bootstrap Aggregation

Exercise 44: Using the Bagging Classifier

Random Forest

Exercise 45: Building the Ensemble Model Using Random Forest

Boosting

Adaptive Boosting

Exercise 46: Adaptive Boosting

Gradient Boosting

Exercise 47: GradientBoostingClassifier

Stacking

Exercise 48: Building a Stacked Model

Activity 14: Stacking with Standalone and Ensemble Algorithms

Summary

Chapter 6

Model Evaluation

Introduction

Exercise 49: Importing the Modules and Preparing Our Dataset

Evaluation Metrics

Regression

Exercise 50: Regression Metrics

Classification

Exercise 51: Classification Metrics

Splitting the Dataset

Hold-out Data

K-Fold Cross-Validation

Sampling

Exercise 52: K-Fold Cross-Validation with Stratified Sampling

Performance Improvement Tactics

Variation in Train and Test Error

Hyperparameter Tuning

Exercise 53: Hyperparameter Tuning with Random Search

Feature Importance

Exercise 54: Feature Importance Using Random Forest

Activity 15: Final Test Project

Summary

Appendix

Chapter 1: Python Machine Learning Toolkit

Activity 1: pandas Functions

Chapter 2: Exploratory Data Analysis and Visualization

Activity 2: Summary Statistics and Missing Values

Activity 3: Visually Representing the Distribution of Values

Activity 4: Relationships Within the Data

Chapter 3: Regression Analysis

Activity 5: Plotting Data with a Moving Average

Activity 6: Linear Regression Using the Least Squares Method

Activity 7: Dummy Variables

Activity 8: Other Model Types with Linear Regression

Activity 9: Gradient Descent

Activity 10: Autoregressors

Chapter 4: Classification

Activity 11: Linear Regression Classifier – Two-Class Classifier

Activity 12: Iris Classification Using Logistic Regression

Activity 13: K-NN Multiclass Classifier

Chapter 5: Ensemble Modeling

Activity 14: Stacking with Standalone and Ensemble Algorithms

Chapter 6: Model Evaluation

Activity 15: Final Test Project