UDA : Unsupervised Data Augmentation

Unofficial PyTorch Implementation of Unsupervised Data Augmentation.

• Experiments on Text Dataset need to be done. Any Pull-Requests would be appreciated.
• Augmentation policies for SVHN, Imagenet using AutoAugment are not available publicly. We use policies from Fast AutoAugment.

Most of codes are from Fast AutoAugment.

Introduction

todo.

Run

$ python train.py -c confs/wresnet28x2.yaml --unsupervised

Experiments

Cifar10 (Reduced, 4k dataset)

Reproduce Paper's Result

SVHN

todo.

ImageNet

todo.

References

• Unsupervised Data Augmentation : https://arxiv.org/abs/1904.12848v1
  • Official Tensorflow Implementation : https://github.com/google-research/uda
• Fast AutoAugment : https://github.com/kakaobrain/fast-autoaugment