Creating a simple Retrieval Augmented Generation (RAG) system using Weaviate and LangChain
This project utilizes language modeling techniques and Weaviate's vector database to retrieve YouTube video content and generate text based on user queries. It incorporates pretrained models and tokenizers from Hugging Face Transformers for text generation and semantic similarity tasks.
Customize this to include your own Youtube Channel!
![Screenshot 2024-07-03 at 5 48 32 PM](https://private-user-images.githubusercontent.com/30791239/345631553-0b9125a2-27d7-4147-bc80-9973bdc28a75.png?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.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.gCArGvUXbCRTqfWatyDuEyuA7VwnN7ym-_iKQRQ-B8c)
- Python 3.6 or higher
- PyTorch
- Transformers from Hugging Face
- Weaviate Python SDK
- Other dependencies listed in
requirements.txt
-
Clone the Repository:
git clone https://github.com/srimoyee1212/VectorRAG.git cd Vector-RAG
-
Install Dependencies
pip install -r requirements.txt
-
Configuration Obtain your Weaviate key from Weaviate Free Trial [https://weaviate.io/]
-
Run your Code
python main.py