alexyhchu / began-tensorflow

👩 Boundary Equilibrium Generative Adversarial Network Implementation in TensorFlow

BEGAN

Prerequisites

• CelebA Dataset
• Tensoflow
• Matplotlib
• Numpy

Getting Started

1. git clone + repo URL
2. cd to repo
3. pip install -r requirement.txt if packages are not yet installed
4. Train model: python train.py downloads dataset if not present and trains the model

Results

Todo

1. Add skip connections Decoder
2. Implement latent space interpolation
3. Test GAN improvement techniques

Built With

• Tensoflow - Software library for numerical computation using data flow graphs
• Matplotlib - Python 2D plotting library
• Numpy - package for scientific computing

Contributing

1. Fork it! Star it?
2. Create your feature branch: git checkout -b my-new-feature
3. Commit your changes: git commit -am 'Add some feature'
4. Push to the branch: git push origin my-new-feature
5. Submit a pull request :D

Authors

• Udacity - Helper functions - Repo
• Jorge Ceja - Model implementations - Account

Acknowledgments

• BEGAN: Boundary Equilibrium Generative Adversarial Networks - arXiv
• Improved Training of Wasserstein GANs - arXiv
• Improved Techniques for Training GANs - arXiv