Mastering Machine Learning with Python in Six Steps : A Practical Implementation Guide to Predictive Data Analytics Using Python

Master machine learning with Python in six steps and explore fundamental to advanced topics, all designed to make you a worthy practitioner. This book’s approach is based on the “Six degrees of separation” theory, which states that everyone and everything is a maximum of six steps away. __Mastering Machine Learning with Python in Six Steps__ presents each topic in two parts: theoretical concepts and practical implementation using suitable Python packages. You’ll learn the fundamentals of Python programming language, machine learning history, evolution, and the system development frameworks. Key data mining/analysis concepts, such as feature dimension reduction, regression, time series forecasting and their efficient implementation in Scikit-learn are also covered. Finally, you’ll explore advanced text mining techniques, neural networks and deep learning techniques, and their implementation. All the code presented in the book will be available in the form of iPython notebooks to enable you to try out these examples and extend them to your advantage. **What You'll Learn*** Examine the fundamentals of Python programming language * Review machine Learning history and evolution * Understand machine learning system development frameworks * Implement supervised/unsupervised/reinforcement learning techniques with examples * Explore fundamental to advanced text mining techniques * Implement various deep learning frameworks **Who This Book Is For** Python developers or data engineers looking to expand their knowledge or career into machine learning area. Non-Python (R, SAS, SPSS, Matlab or any other language) machine learning practitioners looking to expand their implementation skills in Python. Novice machine learning practitioners looking to learn advanced topics, such as hyperparameter tuning, various ensemble techniques, natural language processing (NLP), deep learning, and basics of reinforcement learning. Contents at a Glance......Page 4 Contents......Page 5 About the Author......Page 13 About the Technical Reviewer......Page 14 Acknowledgments......Page 15 Introduction......Page 16 The Best Things in Life Are Free......Page 19 The Rising Star......Page 20 Python 2.7.x or Python 3.4.x?......Page 21 Python from Official Website......Page 22 Python Identifiers......Page 23 Code Blocks (Indentation & Suites)......Page 24 Suites......Page 25 Basic Object Types......Page 26 Comments in Python......Page 28 Multiple Statements on a Single Line......Page 29 Arithmetic Operators......Page 30 Comparison or Relational Operators......Page 31 Assignment Operators......Page 33 Bitwise Operators......Page 34 Membership Operators......Page 36 Identity Operators......Page 37 Selection......Page 38 Iteration......Page 39 Lists......Page 40 Tuple......Page 44 Sets......Page 47 Changing a Set in Python......Page 51 Set Union......Page 52 Set Symmetric Difference......Page 53 Basic Operations......Page 54 Dictionary......Page 55 Defining a Function......Page 60 Scope of Variables......Page 61 Variable Length Arguments......Page 62 Module......Page 63 Opening a File......Page 65 Exception Handling......Page 66 Endnotes......Page 70 Chapter 2: Step 2 – Introduction to Machine Learning......Page 71 History and Evolution......Page 72 Artificial Intelligence Evolution......Page 75 Statistics......Page 76 Bayesian......Page 77 Regression......Page 78 Descriptive Analytics......Page 80 Data Analytics......Page 79 Predictive Analytics......Page 81 Data Science......Page 82 Statistics vs. Data Mining vs. Data Analytics vs. Data Science......Page 84 2) Classification......Page 85 Anomaly Detection......Page 86 Knowledge Discovery Databases (KDD)......Page 87 Preprocessing......Page 88 Cross-Industry Standard Process for Data Mining......Page 89 Phase 6: Deployment......Page 91 Assess......Page 92 KDD vs. CRISP-DM vs. SEMMA......Page 93 Data Analysis Packages......Page 94 Array......Page 95 Creating NumPy Array......Page 96 Field Access......Page 98 Basic Slicing......Page 99 Advanced Indexing......Page 101 Array Math......Page 102 Broadcasting......Page 105 Data Structures......Page 107 Reading and Writing Data......Page 108 Basic Statistics Summary......Page 109 Viewing Data......Page 110 Basic Operations......Page 112 Merge/Join......Page 113 Join......Page 115 Grouping......Page 116 Using Global Functions......Page 118 Customizing Labels......Page 120 Line Plots – Using ax.plot()......Page 121 Multiple Lines on Same Axis......Page 122 Multiple Lines on Different Axis......Page 123 Control the Line Style and Marker Style......Page 124 Line Style Reference......Page 125 Colomaps Reference......Page 126 Bar Plots – using ax.bar() and ax.barh()......Page 127 Horizontal Bar Charts......Page 128 Stacked Bar Example Code......Page 129 Pie Chart – Using ax.pie()......Page 130 Example Code for Grid Creation......Page 131 Machine Learning Core Libraries......Page 132 Endnotes......Page 134 Machine Learning Perspective of Data......Page 135 Nominal Scale of Measurement......Page 136 Ratio Scale of Measurement......Page 137 Feature Engineering......Page 138 Handling Categorical Data......Page 139 Normalizing Data......Page 141 Exploratory Data Analysis (EDA)......Page 143 Univariate Analysis......Page 144 Multivariate Analysis......Page 146 Correlation Matrix......Page 147 Pair Plot......Page 148 Supervised Learning– Regression......Page 149 Correlation and Causation......Page 151 Fitting a Slope......Page 152 R-Squared for Goodness of Fit......Page 154 Mean Absolute Error......Page 156 Polynomial Regression......Page 157 Multivariate Regression......Page 161 Multicollinearity and Variation Inflation Factor (VIF)......Page 163 Interpreting the OLS Regression Results......Page 167 Regression Diagnosis......Page 170 Outliers......Page 171 Homoscedasticity and Normality......Page 172 Over-fitting and Under-fitting......Page 173 Regularization......Page 174 Nonlinear Regression......Page 177 Supervised Learning – Classification......Page 178 Logistic Regression......Page 179 Evaluating a Classification Model Performance......Page 182 ROC Curve......Page 184 Fitting Line......Page 185 Stochastic Gradient Descent......Page 186 Regularization......Page 187 Multiclass Logistic Regression......Page 189 Training Logistic Regression Model and Evaluating......Page 190 Generalized Linear Models......Page 191 Supervised Learning – Process Flow......Page 193 Decision Trees......Page 194 How the Tree Splits and Grows?......Page 195 Conditions for Stopping Partitioning......Page 196 Support Vector Machine (SVM)......Page 198 Key Parameters......Page 199 k Nearest Neighbors (kNN)......Page 201 Components of Time Series......Page 203 Autoregressive Integrated Moving Average (ARIMA)......Page 204 Running ARIMA Model......Page 205 Checking for Stationary......Page 206 Autocorrelation Test......Page 207 Build Model and Evaluate......Page 208 Predicting the Future Values......Page 211 Unsupervised Learning Process Flow......Page 212 K-means......Page 213 Limitations of K-means......Page 214 Elbow Method......Page 217 Average Silhouette Method......Page 219 Key Parameters......Page 221 Principal Component Analysis (PCA)......Page 223 Endnotes......Page 226 Optimal Probability Cutoff Point......Page 227 Rare Event or Imbalanced Dataset......Page 231 Known Disadvantages......Page 234 Which Resampling Technique Is the Best?......Page 235 Variance......Page 236 K-Fold Cross-Validation......Page 237 Ensemble Methods......Page 239 Bagging......Page 240 Feature Importance......Page 242 Extremely Randomized Trees (ExtraTree)......Page 243 How Does the Decision Boundary Look?......Page 244 Boosting......Page 246 Example Illustration for AdaBoost......Page 247 Boosting Iteration 3......Page 248 Final Model......Page 249 Gradient Boosting......Page 251 Boosting – Essential Tuning Parameters......Page 253 Xgboost (eXtreme Gradient Boosting)......Page 254 Ensemble Voting – Machine Learning’s Biggest Heroes United......Page 258 Hard Voting vs. Soft Voting......Page 260 Stacking......Page 262 Hyperparameter Tuning......Page 264 GridSearch......Page 265 RandomSearch......Page 266 Endnotes......Page 268 Chapter 5: Step 5 – Text Mining and Recommender Systems......Page 269 Text Mining Process Overview......Page 270 Data Assemble (Text)......Page 271 Step 2 – Fetching Tweets......Page 273 Sentence Tokenizing......Page 277 Removing Noise......Page 278 Part of Speech (PoS) Tagging......Page 280 Stemming......Page 281 Lemmatization......Page 283 N-grams......Page 285 Bag of Words (BoW)......Page 286 Term Frequency-Inverse Document Frequency (TF-IDF)......Page 288 Frequency Chart......Page 290 Word Cloud......Page 291 Lexical Dispersion Plot......Page 292 Co-occurrence Matrix......Page 293 Outline Placeholder......Page 294 Text Similarity......Page 295 Text Clustering......Page 297 Latent Semantic Analysis (LSA)......Page 298 Latent Dirichlet Allocation (LDA)......Page 300 Text Classification......Page 302 Sentiment Analysis......Page 304 Deep Natural Language Processing (DNLP)......Page 305 Word2Vec......Page 307 Recommender Systems......Page 309 Collaborative Filtering (CF)......Page 310 Endnotes......Page 313 Chapter 6: Step 6 – Deep and Reinforcement Learning......Page 314 Artificial Neural Network (ANN)......Page 315 What Goes Behind, When Computers Look at an Image?......Page 316 Perceptron – Single Artificial Neuron......Page 317 Multilayer Perceptrons (Feedforward Neural Network)......Page 320 Load MNIST Data......Page 321 Key Parameters for scikit-learn MLP......Page 322 Restricted Boltzman Machines (RBM)......Page 324 MLP Using Keras......Page 329 Autoencoders......Page 332 Dimension Reduction Using Autoencoder......Page 333 De-noise Image Using Autoencoder......Page 336 Convolution Neural Network (CNN)......Page 337 CNN on CIFAR10 Dataset......Page 338 CNN on MNIST Dataset......Page 344 Visualization of Layers......Page 348 Recurrent Neural Network (RNN)......Page 349 Long Short-Term Memory (LSTM)......Page 350 Transfer Learning......Page 353 Reinforcement Learning......Page 357 Endnotes......Page 361 Summary......Page 362 Tips......Page 363 Don’t Reinvent the Wheels from Scratch......Page 364 Start with Simple Models......Page 365 Happy Machine Learning......Page 366 Index......Page 367 Exploring fundamental to advanced topics, this book s approach is based on 'Six degrees of separation', which states that everyone and everything is a maximum of six steps away. It also covers advanced text mining techniques, neural networks and deep learning techniques, and their implementation. The code will be available as iPython notebooks.