Skip to content

Production HDC system — integration / comparison ideas? #187

Description

@cedendahlkim

Hi! I'm building a production cognitive system (Oracle Bridge / Bride) that uses HDC as its core similarity engine — 10,000-dimensional binary hypervectors for semantic comparison, anomaly detection, and emotional modeling.

I've been following torchhd for inspiration on the HDC implementation side. My system is written in Rust (for the daemon) with a REST API, and I'd love to explore:

  1. Integration: Could torchhd serve as a Python-side HDC companion? E.g., torchhd for offline batch processing + Oracle Bridge for real-time inference.

  2. Comparison: Has anyone benchmarked torchhd against a running production HDC system? I'd love to compare my binary HV approach against torchhd's implementations.

  3. Cross-pollination: Are there torchhd features that would be useful to expose via a REST API for non-Python clients?

Not trying to sell anything — genuinely interested in the HDC ecosystem. The API has a free tier if you're curious: https://tools.gracestack.se/oracle-bridge-landing.html

Repo with Bride's HDC implementation: https://github.com/cedendahlkim/bride

Cheers!

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions