Skip to content

Extending the vector database article #2

@devayani-kv

Description

@devayani-kv

[Outline] I would like to add to the existing section of vector databases. The general points I would like to add are:

  • How vector databases are different from traditional databases
  • Embeddings (I can just link to the already existing section)
  • A section on the architecture, namely storage, indexing and query processing
  • A primer on Similarity metrics such as Euclidean distance and cosine similarity and an overview of indexing techniques such as HNSW
  • A small tutorial on getting started with an opensource database like ChromaDB, covering
    • various chunking techniques
    • importance of chunk sizes and other hyperparameters related to pushing data into the database.

Metadata

Metadata

Assignees

No one assigned

    Labels

    approvedoutlineOutline of a new article to be submitted

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions