Welcome to the QnA over Structured Data project repository! This project aims to demonstrate the creation of a Question-and-Answer (QnA) system using Large Language Models (LLMs) for structured data sources such as databases and CSV files. The power of LLMs lies in their proficiency in understanding and processing various types of data, making them an excellent tool for extracting valuable insights from structured datasets.
The main objective of this project is to build a QnA system that allows users to ask questions related to the contents of structured data documents. By leveraging LLMs, we can transform raw queries into meaningful interactions with the data, enabling users to retrieve relevant information in a natural language format.
Consider a scenario where you have a structured dataset in the form of a database or a CSV file. You want to enable users to ask questions about the data within these documents without writing complex queries or having deep technical knowledge. This is where the QnA system comes into play. Users can input questions in plain English, and the system, powered by LLMs, will interpret and process these questions to provide accurate and contextual answers based on the underlying data.
To set up and run the QnA system on your local machine, follow these steps:
- Clone the Repository: Begin by cloning this repository to your local environment using the following command:
git clone https://github.com/dwiknrd/llm_qna.git
- Install Dependencies: Navigate to the project directory and install the necessary dependencies:
cd your-repo
pip install -r requirements.txt
make sure, use python=3.10
- Running the QnA System notebook: Execute the QnA system jupyter notebook
Contributions to this project are welcome and encouraged! If you have ideas for improvements, bug fixes, or new features, feel free to submit a pull request. Be sure to follow the established coding style and guidelines.
We would like to express our gratitude to the open-source community for providing the tools and resources that made this project possible.
Happy QnA-ing over your structured data with LLMs! If you have any questions, issues, or suggestions, please don't hesitate to reach out.
To see full version of LLM coursebook you can visit: https://github.com/dwiknrd/llm_coursebook