# Dollar Image Generation using GAN (TensorFlow)
This project demonstrates a simple implementation of a **Generative Adversarial Network (GAN)** using **TensorFlow / Keras** for generating grayscale images resembling dollar images.
The model is trained on a custom image dataset and gradually learns to produce synthetic images through adversarial training.
---
## Features
- Basic GAN architecture (Generator & Discriminator)
- Implemented with TensorFlow 2 and Keras
- Supports custom image datasets
- Periodic image generation during training
- Automatic model checkpoint saving
- Clean and easy-to-read codebase
---
## Project Structure
├── dollar_images/ # Input training images
├── generated_images2/ # Generated samples during training
├── checkpoints/ # Saved model checkpoints
├── train_gan.py # GAN training script
---
## Dataset
All training images should be placed inside the `dollar_images/` directory.
**Dataset details:**
- Supported formats: `.jpg`, `.png`
- Images are automatically:
- Converted to grayscale
- Resized to 28×28
- Normalized to the range [-1, 1]
Example:dollar_images/
├── sample1.jpg
├── sample2.png
├── sample3.jpg
---
## Requirements
Install the required dependencies using pip:
```bash
pip install tensorflow numpy matplotlib pillow
- Place your training images inside the dollar_images/ directory.
- Make sure all required dependencies are installed.
- Run the training script using the following command:
python train_gan.py