Question answering over CSV data with Llama 2

This project explores question answering over CSV data using a pretrained, quantized Llama 2 model. Retrieval supplies relevant document context; the project does not train or fine-tune the model on the uploaded data.

Llama2 CSV ChatBot

It's a conversational tool powered by a fine-tuned large language model (LLM) known as Llama-2 7B. This chatbot utilizes CSV retrieval capabilities, enabling users to engage in multi-turn interactions based on uploaded CSV data.

Suuuuuuu Want to try:

huggingface.co/Llama-2-GGML-CSV-Chatbot

Conversation examples

Streamlit ver. on#v2.0.2.dev20240102

Implementation overview

• Utilizes Llama-2 7B and Sentence Transformers for robust functionality.
• Developed using Langchain and Streamlit technologies for enhanced performance.
• Cross-platform compatibility with Linux, macOS, or Windows OS.

Try it on your

1. Clone the Repository:

git clone https://github.com/ThisIs-Developer/Llama-2-GGML-CSV-Chatbot.git

1. Install Dependencies:

pip install -r requirements.txt

Download the Llama 2 Model:

Download the Llama 2 model file named llama-2-7b-chat.ggmlv3.q4_0.bin from the following link:
Download Llama 2 Model

Llama 2 Model Information

Name	Quant method	Bits	Size	Max RAM required
llama-2-7b-chat.ggmlv3.q4_0.bin	q4_0	4	3.79 GB	6.29 GB

Note: After downloading the model, add the model file to the models directory. The file should be located at models\llama-2-7b-chat.ggmlv3.q4_0.bin, in order to run the code.

Usage

1. Run the Application:

streamlit run app.py

1. Access the Application:
   • Once the application is running, access it through the provided URL.

System Requirements

• CPU: Intel Core i5 or equivalent.
• RAM: 8 GB.
• Disk Space: 7 GB.
• Hardware: Operates on CPU; no GPU required.

Important Notes

• While robust, this chatbot is not a substitute for professional advice.
• Ensure the CSV file adheres to the expected format for optimal performance.