[Feature] --> Add Various Generative Adversarial Network (GAN) Architectures to the GAN/Algorithms/Repository

[UTSAVS26]

Summary:

Introduce various Generative Adversarial Network (GAN) architectures and their applications in medical imaging, with a focus on enhancing diagnostic accuracy and addressing data scarcity and privacy concerns. Additionally, propose the implementation of a basic GAN model alongside Huber Regression to enrich the repository's machine learning toolkit.

Details:

Implementation Tasks:

1. Provide an overview of GAN architectures and their relevance in medical imaging.
2. Develop a new file named BasicGAN.py containing a basic GAN model for generating synthetic medical images.
3. Create unit tests for the Basic GAN in test_BasicGAN.py.
4. Update the README.md file to include:
   • Usage instructions for Basic GAN.
   • Explanation of parameters for model.
   • Installation guide for the repository.
   • Information about running tests.
5. Attach an image illustrating nine different GAN model architectures, including Vanilla GAN, cGANs, EBGAN/BEGAN, InfoGAN, ACGAN, and VAEGAN.
6. Plan to implement each GAN model discussed above with different medical field datasets, enhancing the repository's diversity and utility.

Contributor: Utsav Singhal

Associated Issues: None

Testing Plan:

• Add unit tests for the Basic GAN algorithm, runnable with python -m unittest test_BasicGAN.py.
• Ensure the implementation is tested locally to verify correctness and functionality.

This enhancement aims to provide a comprehensive set of GAN models for medical imaging applications.

Request: Please assign this issue to me under SSOC'24. I intend to add the necessary Python programs to this directory.

Contact Information: LinkedIn: Utsav Singhal

[Avdhesh-Varshney]

@UTSAVS26 go ahead.

[UTSAVS26]

@Avdhesh-Varshney Sorry for the delay in the work. Here's the pull request for this issue raised #55, please review it.