Table of Contents

• Title and Copyright Information
• Preface
  • Audience
  • Documentation Accessibility
  • Related Resources
  • Conventions
• 1 About Oracle Machine Learning for Python
  • 1.1 What Is Oracle Machine Learning for Python?
  • 1.2 Advantages of Oracle Machine Learning for Python
  • 1.3 Transparently Convert Python to SQL
  • 1.4 About the Python Components and Libraries in OML4Py
• 2 Install OML4Py Client for Linux for Use With Autonomous Database on Serverless Exadata Infrastructure
• 3 Install OML4Py for On-Premises Databases
  • 3.1 OML4Py On Premises System Requirements
  • 3.2 Build and Install Python for Linux for On-Premises Databases
  • 3.3 Install the Required Supporting Packages for Linux for On-Premises Databases
  • 3.4 Install OML4Py Server for On-Premises Oracle Database
    • 3.4.1 Install OML4Py Server for Linux for On-Premises Oracle Database 19c
    • 3.4.2 Install OML4Py Server for Linux for On-Premises Oracle Database 21c
    • 3.4.3 Verify OML4Py Server Installation for On-Premises Database
    • 3.4.4 Grant Users the Required Privileges for On-Premises Database
    • 3.4.5 Create New Users for On-Premises Oracle Database
    • 3.4.6 Uninstall the OML4Py Server from an On-Premises Database 19c
  • 3.5 Install OML4Py Client for On-Premises Databases
    • 3.5.1 Install Oracle Instant Client and the OML4Py Client for Linux
      • 3.5.1.1 Install Oracle Instant Client for Linux for On-Premises Databases
      • 3.5.1.2 Install OML4Py Client for Linux for On-Premises Databases
    • 3.5.2 Verify OML4Py Client Installation for On-Premises Databases
    • 3.5.3 Uninstall the OML4Py Client for On-Premises Databases
• 4 Install OML4Py on Exadata
  • 4.1 About Oracle Machine Learning for Python on Exadata
  • 4.2 Configure DCLI to install Python across Exadata compute nodes.
    • 4.2.1 Install Python across Exadata compute nodes using DCLI
    • 4.2.2 Install OML4Py across Exadata compute nodes using DCLI
• 5 Install Third-Party Packages
  • 5.1 Conda Commands
  • 5.2 Administrative Tasks for Creating and Saving a Conda Environment
  • 5.3 OML User Tasks for Downloading an Available Conda Environment
  • 5.4 Using Conda Environments with Embedded Python Execution
• 6 Get Started with Oracle Machine Learning for Python
  • 6.1 Use OML4Py with Oracle Autonomous Database
  • 6.2 Use OML4Py with an On-Premises Oracle Database
    • 6.2.1 About Connecting to an On-Premises Oracle Database
    • 6.2.2 About Oracle Wallets
    • 6.2.3 Connect to an Oracle Database
  • 6.3 Move Data Between the Database and a Python Session
    • 6.3.1 About Moving Data Between the Database and a Python Session
    • 6.3.2 Push Local Python Data to the Database
    • 6.3.3 Pull Data from the Database to a Local Python Session
    • 6.3.4 Create a Python Proxy Object for a Database Object
    • 6.3.5 Create a Persistent Database Table from a Python Data Set
  • 6.4 Save Python Objects in the Database
    • 6.4.1 About OML4Py Datastores
    • 6.4.2 Save Objects to a Datastore
    • 6.4.3 Load Saved Objects From a Datastore
    • 6.4.4 Get Information About Datastores
    • 6.4.5 Get Information About Datastore Objects
    • 6.4.6 Delete Datastore Objects
    • 6.4.7 Manage Access to Stored Objects
• 7 Prepare and Explore Data
  • 7.1 Prepare Data
    • 7.1.1 About Preparing Data in the Database
    • 7.1.2 Select Data
    • 7.1.3 Combine Data
    • 7.1.4 Clean Data
    • 7.1.5 Split Data
  • 7.2 Explore Data
    • 7.2.1 About the Exploratory Data Analysis Methods
    • 7.2.2 Correlate Data
    • 7.2.3 Cross-Tabulate Data
    • 7.2.4 Mutate Data
    • 7.2.5 Sort Data
    • 7.2.6 Summarize Data
  • 7.3 Render Graphics
• 8 OML4Py Classes That Provide Access to In-Database Machine Learning Algorithms
  • 8.1 About Machine Learning Classes and Algorithms
  • 8.2 About Model Settings
  • 8.3 Shared Settings
  • 8.4 Export Oracle Machine Learning for Python Models
  • 8.5 Automatic Data Preparation
  • 8.6 Model Explainability
  • 8.7 Attribute Importance
  • 8.8 Association Rules
  • 8.9 Decision Tree
  • 8.10 Expectation Maximization
  • 8.11 Explicit Semantic Analysis
  • 8.12 Generalized Linear Model
  • 8.13 k-Means
  • 8.14 Naive Bayes
  • 8.15 Neural Network
  • 8.16 Random Forest
  • 8.17 Singular Value Decomposition
  • 8.18 Support Vector Machine
• 9 Automated Machine Learning
  • 9.1 About Automated Machine Learning
  • 9.2 Algorithm Selection
  • 9.3 Feature Selection
  • 9.4 Model Tuning
  • 9.5 Model Selection
• 10 Embedded Python Execution
  • 10.1 About Embedded Python Execution
    • 10.1.1 Comparison of the Embedded Python Execution APIs
  • 10.2 Parallelism with OML4Py Embedded Python Execution
  • 10.3 Embedded Python Execution Views
    • 10.3.1 ALL_PYQ_DATASTORE_CONTENTS View
    • 10.3.2 ALL_PYQ_DATASTORES View
    • 10.3.3 ALL_PYQ_SCRIPTS View
    • 10.3.4 USER_PYQ_DATASTORES View
    • 10.3.5 USER_PYQ_SCRIPTS View
  • 10.4 Python API for Embedded Python Execution
    • 10.4.1 About Embedded Python Execution
    • 10.4.2 Run a User-Defined Python Function
    • 10.4.3 Run a User-Defined Python Function on the Specified Data
    • 10.4.4 Run a Python Function on Data Grouped By Column Values
    • 10.4.5 Run a User-Defined Python Function on Sets of Rows
    • 10.4.6 Run a User-Defined Python Function Multiple Times
    • 10.4.7 Save and Manage User-Defined Python Functions in the Script Repository
      • 10.4.7.1 About the Script Repository
      • 10.4.7.2 Create and Store a User-Defined Python Function
      • 10.4.7.3 List Available User-Defined Python Functions
      • 10.4.7.4 Load a User-Defined Python Function
      • 10.4.7.5 Drop a User-Defined Python Function from the Repository
  • 10.5 SQL API for Embedded Python Execution with On-premises Database
    • 10.5.1 About the SQL API for Embedded Python Execution with On-Premises Database
    • 10.5.2 pyqEval Function (On-Premises Database)
    • 10.5.3 pyqTableEval Function (On-Premises Database)
    • 10.5.4 pyqRowEval Function (On-Premises Database)
    • 10.5.5 pyqGroupEval Function (On-Premises Database)
    • 10.5.6 pyqGrant Function (On-Premises Database)
    • 10.5.7 pyqRevoke Function (On-Premises Database)
    • 10.5.8 pyqScriptCreate Procedure (On-Premises Database)
    • 10.5.9 pyqScriptDrop Procedure (On-Premises Database)
  • 10.6 SQL API for Embedded Python Execution with Autonomous Database
    • 10.6.1 Access and Authorization Procedures and Functions
      • 10.6.1.1 pyqAppendHostACE Procedure
      • 10.6.1.2 pyqGetHostACE Function
      • 10.6.1.3 pyqRemoveHostACE Procedure
      • 10.6.1.4 pyqSetAuthToken Procedure
      • 10.6.1.5 pyqIsTokenSet Function
    • 10.6.2 Embedded Python Execution Functions (Autonomous Database)
      • 10.6.2.1 pyqListEnvs Function (Autonomous Database)
      • 10.6.2.2 pyqEval Function (Autonomous Database)
      • 10.6.2.3 pyqTableEval Function (Autonomous Database)
      • 10.6.2.4 pyqRowEval Function (Autonomous Database)
      • 10.6.2.5 pyqGroupEval Function (Autonomous Database)
      • 10.6.2.6 pyqIndexEval Function (Autonomous Database)
      • 10.6.2.7 pyqGrant Function (Autonomous Database)
      • 10.6.2.8 pyqRevoke Function (Autonomous Database)
      • 10.6.2.9 pyqScriptCreate Procedure (Autonomous Database)
      • 10.6.2.10 pyqScriptDrop Procedure (Autonomous Database)
    • 10.6.3 Asynchronous Jobs (Autonomous Database)
      • 10.6.3.1 oml_async_flag Argument
      • 10.6.3.2 pyqJobStatus Function
      • 10.6.3.3 pyqJobResult Function
      • 10.6.3.4 Asynchronous Job Example
    • 10.6.4 Special Control Arguments (Autonomous Database)
    • 10.6.5 Output Formats (Autonomous Database)
• 11 Administrative Tasks for Oracle Machine Learning for Python
• Index