Vector Databases

• Vector Databases store vectors in higher dimensions
• The text/images are converted into chunks and then into vectors by embeddings.
• Vectors are stored in the vector databases which can now help in similarity search.

Usecases of Vector Databases

• Product search by uploading image
• Recommendation engines
• Facial recognitions

How do Vector Databases search for similar

• Word to vector image
  

• Distance Metrics: These are used to calculate distance between vectors

  • euclidien distance
  • manhattan distance
  • cosine distance
    
• Similiarity Search Approaches:
  • K-NN: K-Nearest Neighbours
  • A-NN: Approximate Nearest Neighbours

Vector Database vs Vector Store

• Vector Databases are purpose built only for vectors
  • Pinecone
  • Weaviate
  • FIASS
  • Chromadb
  • Qdrant
  • Milvus
• Vector Store: This is an existing general purpose database supporting vectors
  • Elastic Search
  • Postgres
  • SQL Server

PineCone

• This is Vector Database as a service offered by Pinecone labs
• This was designed to be cloud native.
• Refer Here to Navigate to pinecone and sigup

Weaviate

• Refer Here for the Weaviate
• Using Weaviate
• Local setup:
  • create a new folder
  • create a file called as docker-compose.yml with contents
  • Now run the command docker compose up -d
  • To remove everything docker compose down -v