As machine learning algorithms become popular, new tools that optimize these algorithms are also developed. Machine Learning Fundamentals explains the scikit-learn API, which is a package created to facilitate the process of building machine learning applications. By explaining the differences between supervised and unsupervised models and by ...

• Preface
• Introduction to Scikit-Learn
  • Introduction
  • Scikit-Learn
    • Advantages of Scikit-Learn
    • Disadvantages of Scikit-Learn
  • Data Representation
    • Tables of Data
    • Features and Target Matrices
    • Exercise 1: Loading a Sample Dataset and Creating the Features and Target Matrices
    • Activity 1: Selecting a Target Feature and Creating a Target Matrix
  • Data Preprocessing
    • Messy Data
    • Exercise 2: Dealing with Messy Data
    • Dealing with Categorical Features
    • Exercise 3: Applying Feature Engineering over Text Data
    • Rescaling Data
    • Exercise 4: Normalizing and Standardizing Data
    • Activity 2: Preprocessing an Entire Dataset
  • Scikit-Learn API
    • How Does It Work?
  • Supervised and Unsupervised Learning
    • Supervised Learning
    • Unsupervised Learning
  • Summary
• Unsupervised Learning: Real-Life Applications
  • Introduction
  • Clustering
    • Clustering Types
    • Applications of Clustering
  • Exploring a Dataset: Wholesale Customers Dataset
    • Understanding the Dataset
  • Data Visualization
    • Loading the Dataset Using Pandas
    • Visualization Tools
    • Exercise 5: Plotting a Histogram of One Feature from the Noisy Circles Dataset
    • Activity 3: Using Data Visualization to Aid the Preprocessing Process
  • k-means Algorithm
    • Understanding the Algorithm
    • Exercise 6: Importing and Training the k-means Algorithm over a Dataset
    • Activity 4: Applying the k-means Algorithm to a Dataset
  • Mean-Shift Algorithm
    • Understanding the Algorithm
    • Exercise 7: Importing and Training the Mean-Shift Algorithm over a Dataset
    • Activity 5: Applying the Mean-Shift Algorithm to a Dataset
  • DBSCAN Algorithm
    • Understanding the Algorithm
    • Exercise 8: Importing and Training the DBSCAN Algorithm over a Dataset
    • Activity 6: Applying the DBSCAN Algorithm to the Dataset
  • Evaluating the Performance of Clusters
    • Available Metrics in Scikit-Learn
    • Exercise 9: Evaluating the Silhouette Coefficient Score and Calinski–Harabasz Index
    • Activity 7: Measuring and Comparing the Performance of the Algorithms
  • Summary
• Supervised Learning: Key Steps
  • Introduction
  • Model Validation and Testing
    • Data Partition
    • Split Ratio
    • Exercise 10: Performing Data Partition over a Sample Dataset
    • Cross Validation
    • Exercise 11: Using Cross-Validation to Partition the Train Set into a Training and a Validation Set
    • Activity 8: Data Partition over a Handwritten Digit Dataset
  • Evaluation Metrics
    • Evaluation Metrics for Classification Tasks
    • Exercise 12: Calculating Different Evaluation Metrics over a Classification Task
    • Choosing an Evaluation Metric
    • Evaluation Metrics for Regression Tasks
    • Exercise 13: Calculating Evaluation Metrics over a Regression Task
    • Activity 9: Evaluating the Performance of the Model Trained over a Handwritten Dataset
  • Error Analysis
    • Bias, Variance, and Data Mismatch
    • Exercise 14: Calculating the Error Rate over Different Sets of Data
    • Activity 10: Performing Error Analysis over a Model Trained to Recognize Handwritten Digits
  • Summary
• Supervised Learning Algorithms: Predict Annual Income
  • Introduction
  • Exploring the Dataset
    • Understanding the Dataset
  • Naïve Bayes Algorithm
    • How Does It Work?
    • Exercise 15: Applying the Naïve Bayes Algorithm
    • Activity 11: Training a Naïve Bayes Model for Our Census Income Dataset
  • Decision Tree Algorithm
    • How Does It Work?
    • Exercise 16: Applying the Decision Tree Algorithm
    • Activity 12: Training a Decision Tree Model for Our Census Income Dataset
  • Support Vector Machine Algorithm
    • How Does It Work?
    • Exercise 17: Applying the SVM Algorithm
    • Activity 13: Training an SVM Model for Our Census Income Dataset
  • Error Analysis
    • Accuracy, Precision, and Recall
  • Summary
• Artificial Neural Networks: Predict Annual Income
  • Introduction
  • Artificial Neural Networks
    • How Do They Work?
    • Understanding the Hyperparameters
    • Applications
    • Limitations
  • Applying an Artificial Neural Network
    • Scikit-Learn's Multilayer Perceptron
    • Exercise 18: Applying the Multilayer Perceptron Classifier Class
    • Activity 14: Training a Multilayer Perceptron for Our Census Income Dataset
  • Performance Analysis
    • Error Analysis
    • Hyperparameter Fine-Tuning
    • Model Comparison
    • Activity 15: Comparing Different Models to Choose the Best Fit for the Census Income Data Problem
  • Summary
• Building Your Own Program
  • Introduction
  • Program Definition
    • Building a Program: Key Stages
    • Understanding the Dataset
    • Activity 16: Performing the Preparation and Creation Stages for the Bank Marketing Dataset
  • Saving and Loading a Trained Model
    • Saving a Model
    • Exercise 19: Saving a Trained Model
    • Loading a Model
    • Exercise 20: Loading a Saved Model
    • Activity 17: Saving and Loading the Final Model for the Bank Marketing Dataset
  • Interacting with a Trained Model
    • Exercise 21: Creating a Class and a Channel to Interact with a Trained Model
    • Activity 18: Allowing Interaction with the Bank Marketing Dataset Model
  • Summary
• Appendix
• Index