Using Oracle Machine Learning with Autonomous Database 2024 ELS

• SECTION I: Introduction to Oracle Machine Learning

Module 1: Oracle Machine Learning Overview

• Cloud-Based Solutions for Analytics
• El Tronics Use Case
• Oracle Autonomous Database
• Basic Machine Learning Flow
• OML Introduction
• Machine Learning: Horizontal Use Cases
• Industry-Specific (Vertical) Use Cases
• Objectives:
• Understand the concept of machine learning in Oracle Cloud.
• Explore industry and horizontal use cases.

Module 2: Features and Components of OML

• OML Components
• Oracle Machine Learning Features
• Steps in the Machine Learning Process
• Objectives:
• Learn the architecture and main components of OML.
• Understand the end-to-end ML workflow.
• SECTION II: Machine Learning Languages in OML

Module 3: Introduction to Machine Learning for Python (OML4Py)

• Features and Advantages
• Using Python for Machine Learning with OML

Module 4: Introduction to Machine Learning for R (OMLR)

• Features and Advantages
• Using R for Machine Learning with OML
• Objectives:
• Compare Python and R ML capabilities in OML.
• Identify scenarios for using OML4Py vs OMLR.
• SECTION III: Workspaces and Projects

Module 5: Creating Workspaces and Projects

• Terry’s Business Problem (Use Case)
• Accessing OML Home Page
• Creating a New Project and Workspace
• Selecting the New Project
• Creating a Workspace

Module 6: Users in Workspaces

• Managing Workspaces
• Workspace Permissions and Types
• Moving Projects Between Workspaces
• Objectives:
• Set up workspaces and projects for ML development.
• Manage user access and permissions effectively.
• SECTION IV: Notebooks and Scratchpad

Module 7: Using the OML Scratchpad Notebook

• Scratchpad Notebook Overview
• Developing Code
• SQL Queries Examples
• Exporting and Importing Notebooks

Module 8: Restrictions on SQL Commands

• Database Options Restrictions
• Database Initialization Parameters
• Modifiable Initialization Parameters
• Objectives:
• Write and test SQL code in scratchpad notebooks.
• Understand limitations and best practices for SQL execution.

Module 9: Introduction to AutoML

• AutoML Concept
• AutoML in OML4Py
• AutoML UI Pipeline
• Algorithm Selection, Feature Selection, Model Tuning

Module 10: AutoML with OML4Py

• Creating DataFrames
• Algorithm Selection and Ranking
• Feature Selection Examples
• Model Tuning and Selection Examples

Module 11: AutoML User Interface (UI)

• AutoML Experiment Pipeline
• Creating AutoML Experiments
• Classification Metrics
• Deploying AutoML Models
• Generating Notebooks from AutoML Models
• Objectives:
• Apply AutoML to automate ML workflows.
• Deploy and generate notebooks from trained models.
• SECTION V: Notebooks and Forms

Module 12: Notebooks

• Creating, Editing, Running Notebooks
• Notebook Examples

Module 13: Forms in Notebooks

• Text Input Forms
• Select Forms
• Check Box Forms
• Output Formats with SET SQLFORMAT

Module 14: Versioning in Notebooks

• Notebook Versions
• Creating and Restoring Versions

Module 15: Templates

• Templates Home Page
• Sharing and Saving Notebooks as Templates
• Personal Templates

Module 16: Instantiating Notebooks from Templates

• Creating Notebooks from Templates
• Sharing Templates and Editing Template Settings
• Example Templates
• Objectives:
• Use notebooks to document, share, and version ML experiments.
• Leverage templates to standardize ML workflows.
• SECTION VI: Jobs and Monitoring

Module 17: Working with Jobs

• Creating Jobs in OML
• Viewing Job Logs
• Job Examples

Module 18: OML Data Monitoring UI

• Value of Data Monitoring
• Data Drift and Metrics
• Creating and Monitoring Data Monitors
• Feature Statistics and Metrics

Module 19: OML Model Monitoring UI

• Value of Model Monitoring
• Concept Drift
• Creating Model Monitors
• Monitoring Metrics, Predictions, and Feature Impact
• Objectives:
• Automate and monitor ML jobs.
• Track data and model performance over time.
• SECTION VII: Administration and Connections

Module 20: Administering Oracle Machine Learning

• Workflow for Managing OML
• User Data Management
• Compute Resource Management
• Creating User Accounts

Module 21: Connection Groups

• (Details TBD based on continuation of content)
• Objectives:
• Administer users, compute resources, and access in OML.
• Set up and manage connection groups for ML projects.

This structure creates 21 clear, learning-focused modules, which follow the logical flow from introduction, setup, development, AutoML, monitoring, and administration.