Fashion MNIST Generation with Various GAN Architectures

Description

This project explores three distinct Generative Adversarial Networks (GAN) architectures for generating images resembling the Fashion MNIST dataset: Normal GANs, DCGANs, and CGANs. Each Jupyter notebook demonstrates a specific implementation, allowing you to easily experiment and compare their capabilities.

Demo

Jupyter Notebooks:

• Normal GAN for Fashion MNIST: Implements the fundamental GAN architecture tailored for the Fashion MNIST dataset.
• DCGAN for Fashion MNIST: Leverages Deep Convolutional GANs (DCGANs) for potential improvements in image quality and stability.
• CGAN for Fashion MNIST: Introduces Conditional GANs (CGANs) with label conditioning, enabling generation of specific fashion items.

Prerequisites

• Python 3.x
• Sklearn
• Flask
• Tensorflow Ensure you have the required dependencies

Usage/Examples

1. Clone this repository:

git clone https://github.com/SanketMagodia/GAN-vs-DCGAN-vs-CGAN.git

2. Open a Jupyter Notebook server:

jupyter notebook

3. Navigate to the project directory in the notebook server.

4. Open and run each Jupyter Notebook individually.

Notes

This is just a test repository.

Contributions

Contributions are welcome! If you have any suggestions, feature requests, or improvements, feel free to open an issue or submit a pull request.