mit-han-lab/data-efficient-gans
[NeurIPS 2020] Differentiable Augmentation for Data-Efficient GAN Training
Implements differentiable augmentation operations (color, translation, cutout) that enable gradient backpropagation through data transformations, allowing discriminators to learn from augmented samples while generators optimize through augmented outputs. Provides portable implementations in both PyTorch and TensorFlow for integration into custom architectures, with pre-configured variants for StyleGAN2, BigGAN, and low-shot scenarios (100-200 images) across CIFAR, FFHQ, and ImageNet datasets.
1,313 stars. No commits in the last 6 months.
Stars
1,313
Forks
176
Language
Python
License
BSD-2-Clause
Category
Last pushed
Sep 24, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/mit-han-lab/data-efficient-gans"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
zhangqianhui/Conditional-GAN
Tensorflow implementation for Conditional Convolutional Adversarial Networks.
EugenHotaj/pytorch-generative
Easy generative modeling in PyTorch
kundan2510/pixelCNN
Theano reimplementation of pixelCNN architecture
pytti-tools/pytti-notebook
Start here
shaohua0116/DCGAN-Tensorflow
A Tensorflow implementation of Deep Convolutional Generative Adversarial Networks trained on...