Machine Learning for Oracle Database Professionals : Deploying Model-Driven Applications and Automation Pipelines

Table of Contents 4 About the Authors 10 About the Technical Reviewer 12 Acknowledgments 13 Introduction 15 Readers and Audiences 15 Chapter 1: Introduction to Machine Learning 17 Why Machine Learning? 17 What Is Machine Learning? 18 Supervised Learning 19 Algorithms for Supervised Learning 23 Unsupervised Learning 25 Algorithms for Unsupervised Learning 28 Semi-Supervised Learning 29 Reinforcement Learning 30 Self-Supervised Learning 32 The Machine Learning Process 32 Summary 38 Chapter 2: Oracle and Machine Learning 39 Oracle Machine Learning for SQL (OML4SQL) 39 Oracle and Other Programming Languages for Machine Learning 40 R 40 Python 41 Java 43 OCI Data Science 43 Oracle Analytics Cloud 48 AutoML 48 Summary 53 Chapter 3: Oracle Machine Learning for SQL 54 PL/SQL Packages for OML4SQL 55 Privileges 55 Data Dictionary Views 56 Predictive Analytics 58 Data Preparation and Transformations 63 Understanding the Data 63 Preparing the Data 66 PL/SQL API for OML4SQL 78 The Settings Table 79 Model Management 81 Model Evaluation 84 Model Scoring and Deployment 93 Partitioned Model 101 Extensions to OML4SQL 104 Oracle Data Miner and Oracle SQL Developer 104 OML Notebooks 110 Summary 110 Chapter 4: Oracle Autonomous Database for Machine Learning 111 Oracle Cloud Infrastructure and Autonomous Database 112 Oracle Cloud Infrastructure Services 113 Sign-up and Access Oracle Cloud Infrastructure 113 Oracle Autonomous Database Architecture and Components 116 Oracle Autonomous Database Attributes 118 Autonomous Database in Free Trier and Always Free 119 Working with Oracle Autonomous Data Warehouse 119 Provisioning Oracle Autonomous Data Warehouse 120 Connect to Oracle Autonomous Data Warehouse 123 Loading Data to Oracle Autonomous Data Warehouse 129 Step 1: Upload a File from a Local Computer to Object Storage 129 Step 2: Create a Credential 132 Step 3: Load Data to a Table in Autonomous Data Warehouse 135 Import Tables/Schema to Oracle Autonomous Database 137 Oracle Machine Learning with ADW 142 Accessing Oracle Machine Learning Through Oracle Autonomous Database 143 Summary 147 Chapter 5: Running Oracle Machine Learning with Autonomous Database 148 Oracle Machine Learning Collaborative Environment 149 Starting with Oracle Machine Learning 149 Sharing Workspaces with Other Users 153 Creating a Machine Learning Notebook 155 Specifying Interpreter Bindings and Connection Groups 156 Running SQL Scripts and Statements 160 Create and Execute SQL Scripts in a Notebook 160 Run SQL Statements in a Notebook 161 Work with Notebooks to Analyze and Visualize Data 163 Summary 167 Chapter 6: Building Machine Learning Models with OML Notebooks 168 Oracle Machine Learning Overview 169 Supervised Learning and Unsupervised Learning 170 Machine Learning Process Flow 173 Oracle Machine Learning for SQL 174 OML4SQL PL/SQL API and SQL Functions 174 Data Preparation and Data Transformation 175 Split Data 175 Data Transformation 175 Transformation Expressions 176 Binning Transformations 177 Model Creation 178 Model Evaluation 180 Model Application 180 Result Comparison 181 Model Scoring and Model Deployment 184 An Example of Machine Learning Project 186 Classification Prediction Example 187 Data Preparation and Data Transformation 188 Predicting Attribute Importance 191 Model Creation 192 Model Testing and Evaluation 194 Model Application 196 Summary 199 Chapter 7: Oracle Analytics Cloud 200 Data Preparation 202 Data Visualization and Narrate 207 Machine Learning in Oracle Analytics Cloud 213 Summary 216 Chapter 8: Delivery and Automation Pipeline in Machine Learning 217 ML Development Challenges 218 Classical Software Engineering vs. Machine Learning 218 Model Drift 218 ML Deployment Challenges 219 ML Life Cycle 220 Scaling Challenges 221 Model Training 221 Model Inference 222 Input Data Processing 222 Key Requirements 222 Design Considerations and Solutions 222 Automating Data Science Steps 223 Automated ML Pipeline: MLOps 224 Model Registry for Tracking 226 Data Validation 229 Pipeline Abstraction 230 Automatic Machine Learning (AutoML) 231 Model Monitoring 231 Model Monitoring Implementation 232 Scaling Solutions 233 ML Accelerators for Large Scale Model Training and Inference 233 Distributed Machine Learning for Model Training 233 Model Inference Options 234 Input Data Pipeline 234 ML Tooling Ecosystem 236 ML Platforms 237 ML Development Tools 238 ML Deployment Tools 239 Summary 239 Chapter 9: ML Deployment Pipeline Using Oracle Machine Learning 240 Mainstream ML Platforms 240 Oracle Machine Learning Environment 243 Data Extraction in Big Data Environment 243 In-Cluster Parallel Data Processing 244 Automated Data Preparation and Feature Engineering 245 General Data Processing Automation 245 Text Processing Automation 246 AutoML 246 Automated Model Selection 246 Automated Feature Selection 246 Automated Hyperparameter Tuning 247 Scalable In-Database Model Training and Scoring 247 In-Database Parallel Execution via Embedded Algorithms 247 Task-Parallel Execution 249 Data-Parallel Execution 249 Degree of Parallelism 249 Environments 249 In-Database Parallel Execution with Partitioned Models 250 In-Cluster Parallel Execution 251 Model Management 252 Saving Models Using R Datastores in Database 252 Leveraging Open Source Packages 255 TensorFlow Extended (TFX) for Data Validation 255 Schema-Based Validation 255 Training and Serving Skew Detection 255 Drift Detection 256 scikit-multiflow for Model Monitoring 256 Kubeflow: Cloud-Native ML Pipeline Deployment 256 Summary 259 Chapter 10: Building Reproducible ML Pipelines Using Oracle Machine Learning 260 The Environment 261 Setting up Oracle Machine Learning for R 262 Verifying the Oracle Machine Learning for R Installation 263 Verifying OML4R on the Server Side 263 Verifying OML4R on the Client Side 264 Setting up Open Source Components 265 The Data 271 Data Validation and Model Monitoring Implementation 272 TensorFlow Data Validation (TFDV) 272 Data Validation 273 Model Monitoring 275 Tracking and Reproducing ML Pipeline 275 Data Version Control (DVC) 276 Versioning Code, Data, and Model Files 276 Demo with Actual ML Pipeline 277 ML Pipeline Project with Git and DVC Initialization and Configuration 278 Defining and Recording Dependencies with DVC 279 Tracking and Reproducing ML Pipelines with DVC 283 Sample Tracking Use Cases 284 Reproducing ML Pipeline: An Example 285 Step 1: Update the Pipeline and Track the Change Using Git 285 Step 2: Reproduce the Pipeline Starting from the Evaluation Stage 285 Separate Storage Locations for Code and Pipeline Artifacts 286 Visualization of ML Pipeline 288 OML4R Troubleshooting Tips 289 Error When Connecting to Oracle Database (as oml_user) 289 Solution 290 Error Due to Missing Packages When Building Models 290 Solution 291 Error When Creating or Dropping R Scripts for Embedded R Execution 292 Solution 293 Summary 293 Index 294 Database developers and administrators will use this book to learn how to deploy machine learning models in Oracle Database and in Oracle's Autonomous Database cloud offering. The book covers the technologies that make up the Oracle Machine Learning (OML) platform, including OML4SQL, OML Notebooks, OML4R, and OML4Py. The book focuses on Oracle Machine Learning as part of the Oracle Autonomous Database collaborative environment. Also covered are advanced topics such as delivery and automation pipelines.Throughout the book you will find practical details and hand-on examples showing you how to implement machine learning and automate deployment of machine learning. Discussion around the examples helps you gain a conceptual understanding of machine learning. Important concepts discussed include the methods involved, the algorithms to choose from, and mechanisms for process and deployment. Seasoned database professionals looking to make the leap into machine learning as a growth path will find much to like in this book as it helps you step up and use your current knowledge of Oracle Database to transition into providing machine learning solutions.What You Will LearnUse the Oracle Machine Learning (OML) Notebooks for data visualization and machine learning model building and evaluationUnderstand Oracle offerings for machine learningDevelop machine learning with Oracle database using the built-in machine learning packagesDevelop and deploy machine learning models using OML4SQL and OML4RLeverage the Oracle Autonomous Database and its collaborative environment for Oracle Machine LearningDevelop and deploy machine learning projects in Oracle Autonomous DatabaseBuild an automated pipeline that can detect and handle changes in data/model performance Who This Book Is ForDatabase developers and administrators who want to learn about machine learning, developers who want to build models and applications using Oracle Database's built-in machine learning feature set, and administrators tasked with supporting applications on Oracle Database that make use of the Oracle Machine Learning feature set Database developers and administrators will use this book to learn how to deploy machine learning models in Oracle Database and in Oracle’s Autonomous Database cloud offering. The book covers the technologies that make up the Oracle Machine Learning (OML) platform, including OML4SQL, OML Notebooks, OML4R, and OML4Py. The book focuses on Oracle Machine Learning as part of the Oracle Autonomous Database collaborative environment. Also covered are advanced topics such as delivery and automation pipelines. Throughout the book you will find practical details and hand-on examples showing you how to implement machine learning and automate deployment of machine learning. Discussion around the examples helps you gain a conceptual understanding of machine learning. Important concepts discussed include the methods involved, the algorithms to choose from, and mechanisms for process and deployment. Seasoned database professionals looking to make the leap into machine learning as a growth path will find much to like in this book as it helps you step up and use your current knowledge of Oracle Database to transition into providing machine learning solutions. What You Will Learn Use the Oracle Machine Learning (OML) Notebooks for data visualization and machine learning model building and evaluation Understand Oracle offerings for machine learning Develop machine learning with Oracle database using the built-in machine[YK1] learning packages Develop and deploy machine learning models using OML4SQL and OML4R Leverage the Oracle Autonomous Database and its collaborative environment for Oracle Machine Learning Develop and deploy machine learning projects in Oracle Autonomous Database Build an automated pipeline that can detect and handle changes in data/model performance Who This Book Is For Database developers and administrators who want to learn about machine learning, developers who want to build models and applications using Oracle Database’s built-in machine learning feature...