---
title: 2 Natural Lang2SQL Query
emoji: 🐠
colorFrom: indigo
colorTo: blue
sdk: gradio
sdk_version: 5.30.0
app_file: app.py
pinned: false
---
# 🧠 Natural Language to SQL Query Generator with Results 🧠

Convert human-readable questions into valid SQL queries using a combination of **semantic search**, **LLMs**, and **Gradio UI** — and see the **query results** live from your database!

## 🚀 Features

* 🔍 **Understand Natural Language:** Converts simple English questions to SQL using powerful embeddings and LLMs (LLaMA3 via Groq).
* 🧠 **Smart Table Matching:** Uses SentenceTransformer-based embeddings and cosine similarity to pick the most relevant tables.
* 🛠️ **Dynamic SQL Generation:** Automatically generates JOINs when needed based on metadata.
* 📄 **Predefined Schema:** Works with three tables — `student`, `employee`, and `course`.
* 🧪 **Instant Execution:** Runs the generated SQL query and shows the result in the UI.
* 🌐 **Interactive UI:** Clean, web-based interface built with Gradio.

---

## 🧰 Tech Stack

* Python 🐍
* [Gradio](https://gradio.app/) — For the frontend UI
* [Sentence Transformers](https://www.sbert.net/) — For encoding metadata and queries
* [Groq API](https://console.groq.com/) — For LLaMA3-powered SQL generation
* SQLite — Lightweight DB backend

---

## 🗃️ Tables Overview

### 🧑‍🎓 `student`

| Column               | Description          |
| -------------------- | -------------------- |
| student\_id          | Unique ID of student |
| first\_name          | First name           |
| last\_name           | Last name            |
| date\_of\_birth      | Birthdate            |
| email                | Email address        |
| phone\_number        | Contact number       |
| major                | Field of study       |
| year\_of\_enrollment | Year enrolled        |

### 👨‍💼 `employee`

| Column            | Description           |
| ----------------- | --------------------- |
| employee\_id      | Unique ID of employee |
| first\_name       | First name            |
| last\_name        | Last name             |
| email             | Email address         |
| department        | Department name       |
| position          | Job title             |
| salary            | Salary                |
| date\_of\_joining | Joining date          |

### 📘 `course`

| Column         | Description                    |
| -------------- | ------------------------------ |
| course\_id     | Unique ID of course            |
| course\_name   | Course title                   |
| course\_code   | Code for the course            |
| instructor\_id | Refers to `employee_id`        |
| department     | Department offering the course |
| credits        | Number of credits              |
| semester       | Semester offered               |

---

## ⚙️ Setup Instructions

### 1. Clone the Repo

```bash
git clone https://github.com/your-username/sql-query-generator.git
cd sql-query-generator
```

### 2. Install Dependencies

```bash
pip install -r requirements.txt
```

### 3. Set Up `.env`

Create a `.env` file in the root directory and add your [Groq API Key](https://console.groq.com/):

```
groq_api_key=your_groq_api_key_here
```

### 4. Launch the App

```bash
python app.py
```

Or just run:

```bash
streamlit run app.py  # If you're using a streamlit version
```

### 5. Use the Web Interface

Go to the URL provided in the terminal (typically `http://127.0.0.1:7860`) and enter your natural language query.

---

## 🧪 Sample Queries

| Natural Language Query                              | Action                        |
| --------------------------------------------------- | ----------------------------- |
| "List all students enrolled in 2022"                | Filters `student` table       |
| "Show all employees in Computer Science department" | Filters `employee` table      |
| "Get courses taught in Fall semester"               | Filters `course` table        |
| "Find course names and instructor names"            | Joins `course` and `employee` |

---

## 🔒 Security & Limitations

* This is a prototype/demo and should not be used in production without input sanitization.
* Currently limited to SELECT queries only.
* Works only with the predefined schema (3 tables).

---

## 🤝 Contributing

Contributions are welcome! Feel free to open an issue or pull request for:

* New table additions
* Better embeddings
* UI/UX improvements
* Model optimization


## 👨‍💻 Author

Built by \Jamgala Bala Swaroop
📫 Email: [jbsroop@gmail.com](mailto:jbsroop@gmail.com)