|
--- |
|
license: mit |
|
language: |
|
- en |
|
pipeline_tag: conversational |
|
tags: |
|
- llama2 |
|
- text-generation |
|
- text-classification |
|
- conversational |
|
library_name: transformers |
|
--- |
|
# 🦙 Llama-2-GGML-CSV-Chatbot |
|
|
|
## Overview |
|
The **Llama-2-GGML-CSV-Chatbot** is 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. |
|
|
|
<img width="2000" src="assets/workflow_1.jpg"> |
|
|
|
## 🚀 Features |
|
|
|
- **CSV Data Interaction:** Allows users to engage in conversations based on uploaded CSV data. |
|
- **Multi-turn Interaction:** Supports seamless multi-turn interactions for a better conversational experience. |
|
|
|
## Development Specs |
|
- Utilizes [Llama-2 7B](https://huggingface.co/TheBloke/Llama-2-7B-Chat-GGML/tree/main) and [Sentence Transformers](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2) for robust functionality. |
|
- Developed using [Langchain](https://github.com/langchain-ai/langchain) and [Streamlit](https://github.com/streamlit/streamlit) technologies for enhanced performance. |
|
- Cross-platform compatibility with Linux, macOS, or Windows OS. |
|
|
|
## 🛠️ Installation |
|
|
|
1. **Clone the Repository:** |
|
```bash |
|
git clone https://github.com/ThisIs-Developer/Llama-2-GGML-CSV-Chatbot.git |
|
``` |
|
2. **Install Dependencies:** |
|
```bash |
|
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](https://huggingface.co/TheBloke/Llama-2-7B-Chat-GGML/tree/main) |
|
|
|
### 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:** |
|
```bash |
|
streamlit run app.py |
|
``` |
|
2. **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. |
|
|
|
## 🤖 How to Use |
|
|
|
- Upon running the application, you'll be presented with a sidebar providing information about the chatbot and an option to upload a CSV file. |
|
- Upload a CSV file containing the data you want the chatbot to interact with. |
|
- Enter your query or prompt in the input field provided. |
|
- The chatbot will process your query and generate a response based on the uploaded CSV data and the Llama-2-7B-Chat-GGML model. |
|
|
|
## 📖 ChatBot Conversession |
|
|
|
### ⚡Streamlit ver. on [#v2.0.2.dev20240102](https://github.com/ThisIs-Developer/Llama-2-GGML-CSV-Chatbot/releases/tag/v2.0.2.dev20240102) |
|
 |
|
|
|
|
|
## 📌 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. |
|
|
|
## 🤝 Contributing |
|
|
|
Contributions and suggestions are welcome! Feel free to fork the repository, make changes, and submit pull requests for improvements or bug fixes. |
|
|
|
## 📄 License |
|
|
|
This project is licensed under the [MIT License](https://github.com/ThisIs-Developer/Llama-2-GGML-CSV-Chatbot/blob/main/LICENSE). |
