Beginning with Machine Learning: The Ultimate Introduction to Machine Learning, Deep Learning, Scikit-learn, and TensorFlow

Cover Page
Title Page
Copyright Page
Dedication Page
About the Authors
About the Reviewer
Acknowledgement
Preface
Errata
Table of Contents
1. Introduction to Machine Learning
    Introduction
    Structure
    Objectives
    What is Machine Learning?
    Why Machine Learning?
    Tools for Machine Learning
        Python
        R language
        Comparison of R and Python
        Golang for ML
        JAVA for ML
        Scikit-learn
        Jupyter Notebook
        Numpy
        Scipy
        Pandas
        Matplotlib
        Integrated Development Environment (IDE)
    Categories of Machine Learning
        Supervised learning
            Classification
            Regression
        Unsupervised learning
        Semi-Supervised Learning (SSL)
        Reinforcement Learning
    Introduction to Scikit-learn
        Characteristics
        Representation of data in scikit-learn
        Challenges encountered
    Conclusion
2. Supervised Learning
    Introduction
    Structure
    Objectives
    Introduction to Supervised Learning
    Generalization
    Overfitting
        Detecting overfitting
        Preventing overfitting
    Underfitting
        Detecting underfitting
        Preventing underfitting
    Bias-Variance Trade-off
    Supervised Machine Learning algorithms
        K-Nearest Neighbors
        Working of k-NN algorithm
        Parameters of KNN Algorithm
        Linear Models
            Regressive type linear models
            Linear models for classification
        Naive Bayes classifiers
            Bayes theorem
            Types of Naive Bayes classifiers
        Decision trees
            Working of decision trees
            Attribute Selection Measure (ASM)
        Random Forests
            Working of Random Forest
            Hyperparameters
        Gradient boosted decision trees
            Hyperparameters
    Conclusion
3. Unsupervised Learning
    Introduction
    Structure
    Objectives
    Clustering
        Types of clustering
        Clustering algorithms
        K-Means clustering algorithm
            Working of the K-Means algorithm
            Evaluation methods
            Elbow method
            Applications
            Drawbacks
        Hierarchical clustering
            Agglomerative hierarchical clustering
            Divisive hierarchical clustering
    Dimensionality reduction
        The curse of dimensionality
        Approaches towards dimensionality reduction
            Feature selection
            Feature extraction
            Linear dimensionality reduction methods
            Non-linear dimensionality reduction methods
    Conclusion
4. Model Evaluation
    Introduction
    Structure
    Objectives
    Cross-validation
        Non-exhaustive method
            Holdout method
            K-fold cross-validation
            Stratified K-fold Cross-Validation
            Advantages of Cross-Validation
            Disadvantages of Cross-Validation
    Evaluation metrics
        Metrics of classification
            Classification Accuracy
            Confusion Matrix
            Precision
            Recall
            F1 Score
            ROC Curve
            Precision-Recall Curve
    Metrics of Regression
        R Squared
        Mean Square Error
        Root Mean Square Error
        Mean Absolute Error
    Conclusion
5. Reinforcement Learning
    Introduction
    Structure
    Objectives
    Basics of Reinforcement Learning
    Policy
        OpenAI gym
        Policy optimization
        Policy Gradient (PG)
    Markov Decision Process (MDP)
        Markov property
        Markov Chain
        Markov Reward Process
        Markov Decision Process (MDP)
        Return
        Discount
        Policy
        Value Function
        Bellman Expectation Equation
        State Value Function
        Action value function
        Bellman Expectation Equation
        Optimal Value Function
        Bellman Optimality Equation
    Temporal Difference Learning
        TD(1) Algorithm
        TD(0) Algorithm
        TD(λ) Algorithm
    Q Learning
        Working of the algorithm
    Approximate Q Learning
        Advantages
        Disadvantages
    Conclusion
6. Neural Networking and Deep Learning
    Introduction
    Structure
    Objectives
    Introduction to TensorFlow
        Low-Level APIs
        High-Level APIs
    TensorFlow basics
    Artificial Neural Networks (ANN)
    History of Artificial Neural Networks
    Biological neurons
    How is an ANN trained?
    Computations with neuron
    Perceptron
    Activation functions
        ReLU
        Sigmoid function
        Hyperbolic Tangent Function (tanh)
    Multi-layer perceptron
    Backpropagation algorithm
    Fine-tuning of neural networks
    Number of hidden layers
    Activation function
    Number of Neurons
    Convolutional Neural Networks
    Convolution
    Pooling layer
    Architecture of CNN
    Fully connected layer
    Dropout
    Recurrent Neural Networks (RNN)
    Recurrent neurons
    Backpropagation in RNN
    Types of RNN
        One-to-one
        One-to-many
        Many-to-one
        Many-to-many
        Many-to-many
    Problems faced by RNNs
        Exploding gradients
        Vanishing gradients
    Solutions to these problems
        Reduce the number of layers
        Gradient clipping
        Weight initialization
        LSTM networks
    Long Short Term Memory (LSTM)
        Working of LSTM cells
        Sigmoid activation
        Forget gate
        Input gate
        Output gate
    Gated Recurrent Unit (GRU)
        Update gate
        Reset gate
    Conclusion
7. Appendix: Machine Learning Questions
    Chapter 1: Introduction to Machine Learning
        Extra Questions: Chapter 1
    Chapter 2: Supervised Learning
    Chapter 3: Unsupervised Learning
    Chapter 4: Model Evaluation
    Chapter 5: Reinforcement Learning
    Chapter 6: Neural Networks and Deep Learning
    MCQ Questions
        Question links with answers
    Some Other Questions
    True and False Questions
    Fill in the blanks
    Extra Questions: Mix
Index