README.md

metadata

title: Semantic Search Cache API
emoji: 🔎
colorFrom: blue
colorTo: purple
sdk: docker
python_version: '3.10'
app_file: app.py
app_port: 7860
pinned: false

Semantic Search Cache API

This project provides a semantic search engine with caching and fuzzy clustering.

It uses:

• Sentence embeddings
• FAISS vector index
• Gaussian Mixture Model clustering
• Semantic caching

The API is built with FastAPI and deployed using Docker on Hugging Face Spaces.

---

Features

• Semantic similarity search
• FAISS vector indexing
• Semantic cache for faster repeated queries
• Fuzzy clustering with GMM
• REST API endpoints

---

API Endpoints

POST /query

Example request:

{
  "query": "space shuttle launch",
  "top_k": 5
}

GET /cache/stats

Returns semantic cache statistics.

DELETE /cache

Clears the cache.

---

Project Structure

semantic-search-cache/
│
├── api/
│   └── main.py
│
├── src/
│   ├── query_engine.py
│   ├── semantic_cache.py
│   ├── fuzzy_cluster.py
│
├── models/
│   ├── faiss_index.index
│   ├── documents.pkl
│   └── gmm_model.pkl
│
├── requirements.txt
├── Dockerfile
└── app.py

---

Deployment

This project is deployed using Docker on Hugging Face Spaces.

The container exposes port 7860 and runs a FastAPI server using Uvicorn.