Course :: Improve search performance in .NET app with Vector Search

• 2 days

Software requirements

• .NET 8
• VS Code
• Docker
  • Ollama
  • Qdrant
  • PostgreSQL

Outline

• Introduction to Vector search

  • Vector data types
  • Distance algorithm (similarity search)
    • Euclidean distance
    • Cosign distance
  • Embedding data (convert text to machine-readable code)
    • Multilingual embedding
    • Embedding models

• Introduction to Vector Database

  • Install Qdrant with Docker
  • Working with Qdrant
    • CRUD operation (Create/REad,Update/Delete)
    • Search operations
      • keyword
      • vector/semantic
      • hybrid

• Working with Hybrid search

  • Keyword search or Full-text search
  • Vector search or Semantic search

• Local LLM with Ollama

  • Introduction to Ollama
  • Install and configuration Ollama with Docker
  • Install embedding model
    • bge-m3
  • Workshop with Ollama
    • Create embedding data (vector) from text via REST API

• Workshop with .NET 8 (Improve search data)

  • Create Project with web-api project
  • Create sample data into PostgreSQL database
  • Generate embedding data from sample data with Ollama API
  • Insert embedding data to Vector database (Qdrant) with Qdrant API
  • Search data from Vector database (Qdrant) with Qdrant API
    • keyword
    • vector/semantic
    • hybrid