AuthorGAN: Improving GAN Reproducibility using a Modular GAN Framework

TL;DR

AuthorGAN is a modular, library-agnostic framework with a visual designer that simplifies GAN creation, aiming to democratize access and reduce the expertise barrier for using GAN models.

Contribution

It introduces a novel, modular, library-agnostic representation of GANs and a drag-and-drop visual interface for easy architecture design, enhancing accessibility.

Findings

Demonstrated interoperability across Keras, TensorFlow, and PyTorch.

Implemented five GAN models within the framework.

Evaluated models on the MNIST dataset.

Abstract

Generative models are becoming increasingly popular in the literature, with Generative Adversarial Networks (GAN) being the most successful variant, yet. With this increasing demand and popularity, it is becoming equally difficult and challenging to implement and consume GAN models. A qualitative user survey conducted across 47 practitioners show that expert level skill is required to use GAN model for a given task, despite the presence of various open source libraries. In this research, we propose a novel system called AuthorGAN, aiming to achieve true democratization of GAN authoring. A highly modularized library agnostic representation of GAN model is defined to enable interoperability of GAN architecture across different libraries such as Keras, Tensorflow, and PyTorch. An intuitive drag-and-drop based visual designer is built using node-red platform to enable custom architecture designing without the need for writing any code. Five different GAN models are implemented as a part of this framework and the performance of the different GAN models are shown using the benchmark MNIST dataset.