Oracle AI Database: AI Vector Search Training

Module 1: Overview of Oracle AI Vector Search

• Objectives
• VECTOR Data Type
• Vector Embeddings
• Similarity Search
• Vector Embedding Models
• Summary

Module 2: Why Use Oracle AI Vector Search

• Objectives
• Benefits of Oracle AI Vector Search
• Examples
• Summary

Module 3: Oracle AI Vector Search Workflow

• Objectives
• Generate Vector Embeddings
• Embedding Generation Examples
• Load ONNX Model into the DB
• Generate the Embedding Example
• Store Vector Embeddings
• Vector Indexes
• Query Data with Similarity Searches
• Complete Workflow
• Summary

Module 4: Running Basic Queries on Vectors

• Objectives
• Basic Queries & Comparison Operations
• Demo
• Summary

Module 5: Vector Indexes and Memory

• Objectives
• Vector Indexes Overview
• Why Vector Indexes are Needed
• Vector Pool
• In-Memory Neighbor Graph Vector Index (HNSW)
• Neighbor Partition Vector Index (IVF)
• Creating a Basic Vector Index & Important Parameters
• Using Vector Indexes & Monitoring Accuracy
• Important Limitations & Best Practices
• Memory Considerations: Vector Storage & In-Memory Indexes
• Sample Calculation
• Summary

Module 6: DML Operations on Vectors

• Objectives
• Create a Table with a Vector Column
• VECTOR Data Type Declaration Formats
• Vector DML Operations
• Vector DML Using SQL*Loader
• Demo
• Summary

Module 7: Vector DDL

• Objectives
• Tables with Different Vector Formats
• Insert Examples
• DDL Operations on Vectors
• Prohibited Operations & VECTOR Data Type Restrictions
• Demo
• Summary

Module 8: Creating and Finding the Nearest Vectors

• Objectives
• Vector Constructor & Examples
• Vector Distance & Operands

Distance Metrics: Euclidean, Euclidean Squared, Cosine, Dot Product, Manhattan, Hamming

• Vector Distance Examples & Shorthand Operators
• Demo
• Summary

Module 9: Finding the Closest Vectors

• Objectives
• Exact Similarity Search (Euclidean & Euclidean Squared)
• Approximate Similarity Search Overview
• HNSW & IVF Indexes
• Multi-Vector Similarity Search
• Demo
• Summary

Module 10: Narrowing Search Results

• Objectives
• Attribute Filtering
• Demo
• Summary

Module 11: Testing Other Distance Functions

• Objectives
• L1_DISTANCE, L2_DISTANCE, COSINE_DISTANCE, INNER_PRODUCT
• Demo
• Summary

Module 12: Testing Other Vector Functions

• Objectives
• Vector Constructors
• Vector Serializers & Norm
• Vector Dimension Count & Format
• Demo
• Summary

Module 13: RAG Overview

• Agenda & Topics
• Vector Data Workflow
• Retrieval Augmented Generation (RAG) Workflow
• Interacting with LLMs
• RAG with Oracle AI Vector Search
• Summary

Module 14: Using Embedding Models with Oracle AI Vector Search

• Agenda & Topics
• Embedding Models Overview
• VECTOR_EMBEDDING()
• Create Table & Vectorize a Table
• Similarity Search
• Change Embedding Models
• Summary

Module 15: Oracle Vector Search & OCI Generative AI Service (Python)

• Agenda & Topics
• OCI Gen Service Overview
• Summary of Steps: Load Sources, Text Chunks, Vectorize, Create and Call
• Summary

Module 16: RAG with Oracle AI Vector Search & OCI Gen AI Service (PL/SQL)

• Agenda & Topics
• Process Overview
• Step-by-Step: Steps 1–5
• Invoke the Chain
• Summary

Module 17: Oracle AI Vector Search Supporting Features

• Agenda & Topics
• Exadata AI Storage
• GoldenGate Microservices for AI
• Distributed AI Processing & Real-Time Vector Hub
• SQL Loader & Data Pump
• Summary

Module 18: Select AI with Autonomous

• Agenda & Topics
• Select AI Overview
• Benefits & Challenges
• Using SQL for Select AI Queries
• Summary