|
|
--- |
|
|
title: Multi-Source RAG Assistant |
|
|
emoji: 📚 |
|
|
colorFrom: indigo |
|
|
colorTo: blue |
|
|
sdk: streamlit |
|
|
sdk_version: 1.45.0 |
|
|
app_file: app.py |
|
|
pinned: false |
|
|
--- |
|
|
|
|
|
# 📚 Multi-Source RAG Assistant |
|
|
|
|
|
This app lets users interact with: |
|
|
- 🧾 PDF documents |
|
|
- 📊 CSV datasets (with automatic EDA) |
|
|
- 🌐 Any website (scraped and embedded into a vector database) |
|
|
|
|
|
💡 Powered by: |
|
|
- Google Gemini API |
|
|
- FAISS vector search |
|
|
- LangChain framework |
|
|
|
|
|
## 🔧 How It Works |
|
|
|
|
|
1. Select input type from sidebar: **PDF**, **CSV**, or **Website URL**. |
|
|
2. Upload or input accordingly. |
|
|
3. Ask questions — the assistant answers using context-aware RAG via Gemini. |
|
|
|
|
|
## 🚀 Getting Started |
|
|
|
|
|
Before using the app: |
|
|
- Enter your **Gemini API Key** in the sidebar. |
|
|
|
|
|
## 🧠 Tech Stack |
|
|
|
|
|
- `Streamlit` for UI |
|
|
- `LangChain` for RAG logic |
|
|
- `FAISS` for vector storage |
|
|
- `SentenceTransformers` for embeddings |
|
|
- `Google Generative AI (Gemini)` for LLM-powered answers |
|
|
|
|
|
--- |
|
|
|
|
|
### 🛠 Developed by Abhijeet Singh |
|
|
|
|
|
Hosted with ❤️ on [Hugging Face Spaces](https://huggingface.co/spaces) |
