As the name suggests this is a Text-to-SQL chatbot which users can now clone and operate on their own local systems, utilizing MySQL Server and Gogle Gemini API Key.
The project allows users to ask questions in natural language, which are then converted into SQL queries using Google GenerativeAI model and LangChain. This bot then further executes the queries on the MySQL database to generate accurate responses.
In my demo I have made use of an Uber Dataset with the given ER Diagram shown below.
When configured and deployed this is what it finally produces:
Techstack used:
Google GenerativeAI LLM: For natural language processing and query generation. LangChain's SQLDatabaseChain: To integrate the model with the MySQL database. Hugging Face Embeddings: For enhanced semantic understanding. Chromadb: As a vector store for efficient data retrieval. Few-Shot Learning: Implemented to fine-tune the model for more precise results. Streamlit: For building a user-friendly interface.