CoolFace
Apppublic

maheshsmc/dense-retriever-day6

sourceHugging Faceupdated 1y agoView on Hugging Face
0likes
App README

Dense Retriever — Day 6 (Embeddings vs Keywords)

This Space demonstrates a dense (embedding) retriever using SentenceTransformers. Paste your own small corpus (one passage per line) or use the default tiny corpus.

How to use

  1. 1.Enter a query.
  2. 2.(Optional) Paste a custom corpus (one line = one doc). Leave empty to use the default.
  3. 3.Choose Top‑K.
  4. 4.Click Submit.

Local run

bash
pip install -r requirements.txt
python app.py

Notes

  • —Model: all-MiniLM-L6-v2 (fast & lightweight)
  • —CPU works fine for tiny demos.