cache models for faster startup

Dockerfile

@@ -4,8 +4,14 @@ WORKDIR /app
 
 COPY . /app
 
+# Install dependencies
 RUN pip install --no-cache-dir -r requirements.txt
 
+# Pre-download models so they are cached in the image
+RUN python -c "from sentence_transformers import SentenceTransformer; SentenceTransformer('sentence-transformers/all-MiniLM-L6-v2')"
+
+RUN python -c "from sentence_transformers import CrossEncoder; CrossEncoder('cross-encoder/ms-marco-MiniLM-L-6-v2')"
+
 EXPOSE 7860
 
 CMD ["uvicorn","main:app","--host","0.0.0.0","--port","7860"]

README.md

@@ -1,10 +1,119 @@
----
-title: Scholar Rag Engine
-emoji: 🚀
-colorFrom: gray
-colorTo: red
-sdk: docker
-pinned: false
----
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+
+# Scholar RAG Engine
+
+Scholar RAG Engine is a Retrieval-Augmented Generation (RAG) system designed for answering questions from PDFs and web pages.
+
+The system extracts content, builds semantic indexes, retrieves relevant context, and generates answers using an LLM.
+
+## Features
+
+- PDF document indexing
+- Website content scraping
+- Hybrid semantic retrieval
+- ColBERT-style retrieval
+- Cross-encoder reranking
+- LLM answer generation
+- Modern UI with dark mode
+- Expandable retrieved context viewer
+
+## Architecture
+
+Pipeline:
+
+User Query
+↓
+Retriever (ColBERT)
+↓
+Reranker (Cross Encoder)
+↓
+Context Compression
+↓
+LLM (Gemini)
+↓
+Final Answer
+
+## Tech Stack
+
+Backend:
+- FastAPI
+- Python
+
+Retrieval:
+- Sentence Transformers
+- FAISS
+- ColBERT-style token similarity
+
+Ranking:
+- Cross Encoder (MS MARCO)
+
+LLM:
+- Google Gemini API
+
+Frontend:
+- HTML
+- CSS
+- JavaScript
+
+Deployment:
+- Hugging Face Spaces
+- Docker
+
+## Project Structure
+scholar-rag-engine
+│
+├── main.py
+├── ingestion.py
+├── chunking.py
+├── scraper.py
+├── retrieval_colbert.py
+├── reranker.py
+├── LLM.py
+├── requirements.txt
+├── Dockerfile
+│
+└── templates
+└── index.html
+
+
+## Installation
+
+Clone the repository
+git clone https://github.com/mr-snake-mr/scholar-rag-engine
+cd scholar-rag-engine
+
+
+Install dependencies
+pip install -r requirements.txt
+
+
+Run the server
+uvicorn main:app --reload
+
+
+Open in browser
+http://localhost:8000
+
+
+## Environment Variables
+
+Set your Gemini API key:
+GOOGLE_API_KEY=your_gemini_api_key
+
+
+## Deployment
+
+This project is deployed on Hugging Face Spaces using Docker.
+https://huggingface.co/spaces/snakeeee/scholar-rag-engine
+
+
+## Future Improvements
+
+- Streaming responses
+- Chat-style UI
+- Multi-document support
+- Vector database integration
+- GPU acceleration
+
+## Author
+
+Developed as an AI-powered research assistant project.