sony/san
PyTorch implementation of slicing adversarial network (SAN)
This project helps machine learning researchers and practitioners improve the performance of their Generative Adversarial Networks (GANs). By applying the Slicing Adversarial Network (SAN) modifications, you can get higher-quality images or data generated by your GAN models. You input an existing GAN model architecture and training data, and it outputs a modified GAN that achieves better performance, specifically in metrics like FID scores for image generation.
Use this if you are a machine learning researcher or engineer working with GANs and want to improve the quality and realism of the synthetic data or images your models generate.
Not ideal if you are looking for an out-of-the-box data generation tool without needing to delve into GAN architectures or training methodologies.
Stars
98
Forks
7
Language
Python
License
—
Category
Last pushed
Dec 01, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/sony/san"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
huggingface/pytorch-pretrained-BigGAN
🦋A PyTorch implementation of BigGAN with pretrained weights and conversion scripts.
torchgan/torchgan
Research Framework for easy and efficient training of GANs based on Pytorch
metal3d/keras-video-generators
Keras generators to generate sequences from videos as input
GANs-in-Action/gans-in-action
Companion repository to GANs in Action: Deep learning with Generative Adversarial Networks
junyanz/pytorch-CycleGAN-and-pix2pix
Image-to-Image Translation in PyTorch