---
title: Decomate
emoji: ⚡
colorFrom: red
colorTo: blue
sdk: gradio
sdk_version: 5.38.2
app_file: app.py
pinned: false
license: cc-by-nc-3.0
---

## Overview

Decomate is a Gradio-based application designed for intelligent SVG decomposition and animation generation. It leverages advanced machine learning models to analyze SVG files, group elements semantically, and create animations based on user descriptions.

## Features

- **SVG Decomposition**: Automatically parse and decompose SVG files into semantic groups such as characters, objects, and backgrounds.
- **Animation Suggestions**: Generate animation ideas tailored to the SVG structure.
- **Interactive UI**: User-friendly interface for uploading SVGs, analyzing them, and previewing animations.
- **Custom Animation**: Create animations using AI-driven suggestions and user inputs.

## Installation

1. Clone the repository:

   ```bash
   git clone https://github.com/your-repo/decomate.git
   cd decomate
   ```

2. Install dependencies:

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

3. Set up environment variables:
   - Create a `.env` file in the root directory.
   - Add your `ANTHROPIC_API_KEY`:
     ```
     ANTHROPIC_API_KEY=your_api_key_here
     ```

## Usage

1. Launch the application:

   ```bash
   python app.py
   ```

2. Open the Gradio interface in your browser.

3. Upload an SVG file or paste SVG markup to start analyzing and animating.

## File Structure

- **`app.py`**: Main application file.
- **`utils.py`**: Utility functions for SVG processing.
- **`requirements.txt`**: List of dependencies.
- **`prompt_template/`**: Templates for AI prompts.
- **`dataset/`**: Sample SVG files for testing.
- **`.env`**: Environment variables (not included in the repository).

## License

This project is licensed under the Creative Commons Attribution-NonCommercial 3.0 License. See the `LICENSE` file for details.

## Acknowledgments

- **Gradio**: For providing an intuitive interface for machine learning applications.
- **Anthropic**: For enabling advanced language model capabilities.
- **CairoSVG**: For SVG to PNG conversion.