khoj/tests/data/notes/main_readme.org

• Khoj
  • Dependencies
  • Install
  • Run
  • Use
  • Acknowledgments

Khoj

Allow natural language search on user content like notes, images using transformer based models

All data is processed locally. User can interface with khoj app via Emacs, API or Commandline

Dependencies

• Python3
• Miniconda

Install

git clone https://github.com/debanjum/khoj && cd khoj
conda env create -f environment.yml
conda activate khoj

Run

Load ML model, generate embeddings and expose API to query specified org-mode files

python3 main.py --input-files ~/Notes/Schedule.org ~/Notes/Incoming.org --verbose

Use

• Khoj via Emacs

  • Install khoj.el
  • Run M-x khoj <user-query> or Call C-c C-s

• Khoj via API

  • Query: GET http://localhost:8000/search?q="What is the meaning of life"
  • Regenerate Embeddings: GET http://localhost:8000/regenerate
  • Khoj API Docs

• Call Khoj via Python Script Directly

  python3 search_types/asymmetric.py \
  --compressed-jsonl .notes.jsonl.gz \
  --embeddings .notes_embeddings.pt \
  --results-count 5 \
  --verbose \
  --interactive

Acknowledgments

• MiniLM Model for Asymmetric Text Search. See SBert Documentation
• OpenAI CLIP Model for Image Search. See SBert Documentation
• Charles Cave for OrgNode Parser