Skip to content

README.md

Latest commit

 

History

History
123 lines (82 loc) · 2.91 KB

README.md

File metadata and controls

123 lines (82 loc) · 2.91 KB

📊 AI Data Visualization Agent (Gemini + Local Execution)

This Streamlit app allows you to analyze and visualize datasets using AI. Simply upload a CSV file and ask natural language questions — Google’s Gemini model will automatically generate and execute Python code locally to produce insights and plots.

🚀 Features

✅ Upload any CSV file ✅ Ask natural questions like “Compare average age by gender” or “Show correlation heatmap” ✅ Gemini (free API) generates runnable Python code ✅ Executes safely on your machine — no data is sent to any external server ✅ Supports matplotlib, seaborn, and plotly for visualizations ✅ Displays AI-generated charts, tables, and code outputs interactively

🧠 How It Works

  1. You upload a dataset (e.g., titanic.csv).
  2. You type a query (e.g., “Group by Pclass and show survival rate”).
  3. The Gemini API generates Python analysis code.
  4. The app executes the generated code locally using your df DataFrame.
  5. Results and visualizations appear right in the Streamlit dashboard.

🧪 Tech Stack

  • Frontend: Streamlit
  • LLM: Google Gemini API (gemini-2.0-flash)
  • Visualization: Matplotlib, Seaborn, Plotly
  • Data: Pandas

🛠️ Setup & Installation

1. Clone the repository

git clone https://github.com/your-username/ai-data-visualization-agent.git
cd ai-data-visualization-agent

2. Create a virtual environment (optional but recommended)

python -m venv venv
source venv/bin/activate   # on macOS/Linux
venv\Scripts\activate      # on Windows

3. Install dependencies

pip install -r requirements.txt

4. Get a free Gemini API key

  • Visit Google AI Studio
  • Create a free API key
  • Copy it — you’ll need it inside the app’s sidebar.

5. Run the Streamlit app

streamlit run app.py

🧩 Example Usage

Example Query:

“Group by ‘Pclass’ and show average fare and age using a bar chart.”

Output:

  • Gemini generates Python code using Pandas & Matplotlib.
  • The code runs locally.
  • The app displays a bar chart comparing average age and fare per passenger class.

🧠 Notes & Safety

  • This app executes AI-generated code locally on your machine. Avoid running it on shared or public servers.
  • Your data and API key remain private — no data is uploaded externally.
  • Works best with well-formatted CSV datasets.

📦 Requirements

All dependencies are listed in requirements.txt:

streamlit>=1.37.0
pandas>=2.0.0
matplotlib>=3.7.0
seaborn>=0.12.2
plotly>=5.15.0
Pillow>=10.0.0
google-generativeai>=0.7.2

💡 Future Improvements

  • Add AI-generated data summaries
  • Export analysis reports as PDF
  • Support for Excel & JSON files
  • Memory for multi-step queries

⭐ If you like this project, consider starring the repo!