TorchGAN: A Flexible Framework for GAN Training and Evaluation
This provides a practical tool for researchers and practitioners working with GANs, though it is incremental as it builds on existing PyTorch infrastructure.
The authors tackled the need for a flexible and efficient framework for GAN training and evaluation by developing TorchGAN, a PyTorch-based tool that enables succinct code and modular customization with zero overhead compared to baseline implementations.
TorchGAN is a PyTorch based framework for writing succinct and comprehensible code for training and evaluation of Generative Adversarial Networks. The framework's modular design allows effortless customization of the model architecture, loss functions, training paradigms, and evaluation metrics. The key features of TorchGAN are its extensibility, built-in support for a large number of popular models, losses and evaluation metrics, and zero overhead compared to vanilla PyTorch. By using the framework to implement several popular GAN models, we demonstrate its extensibility and ease of use. We also benchmark the training time of our framework for said models against the corresponding baseline PyTorch implementations and observe that TorchGAN's features bear almost zero overhead.