Skip to content

Multi block + Random Erasing Data Augmentation. Experiments on CIFAR10, CIFAR100 and Fashion-MNIST

License

Notifications You must be signed in to change notification settings

hyunjiLee123/Multi-block-Random-Erasing-Frequency-domain

 
 

Repository files navigation

Multi block + Random Erasing in the Frequency domain

===============================================================

This code has the source code for the paper
"멀티 블록 기반 Random Erasing in the Frequency Domain".

Installation

Requirements for Pytorch (see Pytorch installation instructions)

Dataset

You should download 'cifar10-c', 'cifar100-c', 'fmnist-c' dataset if you want to evaluate corruption error

Examples:

CIFAR10

ResNet-20 baseline on CIFAR10: python cifar.py --dataset cifar10 --arch resnet --depth 20

ResNet-20 + Random Erasing on CIFAR10: python cifar.py --dataset cifar10 --arch resnet --depth 20 --p 0.5

ResNet-20 + REF on CIFAR10: python cifar.py --dataset cifar10 --arch resnet --depth 20 --p 0.0 --p2 0.5

CIFAR100

ResNet-20 baseline on CIFAR100: python cifar.py --dataset cifar100 --arch resnet --depth 20

ResNet-20 + Random Erasing on CIFAR100: python cifar.py --dataset cifar100 --arch resnet --depth 20 --p 0.5

ResNet-20 + REF on CIFAR100: python cifar.py --dataset cifar100 --arch resnet --depth 20 --p 0.0 --p2 0.5

Fashion-MNIST

ResNet-20 baseline on Fashion-MNIST: python fashionmnist.py --dataset fashionmnist --arch resnet --depth 20

ResNet-20 + Random Erasing on Fashion-MNIST: python fashionmnist.py --dataset fashionmnist --arch resnet --depth 20 --p 0.5

Other architectures

For ResNet: --arch resnet --depth (20, 32, 44, 56, 110)

For WRN: --arch wrn --depth 28 --widen-factor 10

===============================================================

Corruption data preparation

CIFAR-10-C

Download CIFAR10-C dataset

https://paperswithcode.com/dataset/cifar-10c

CIFAR-100-C

Download CIFAR100-C dataset

https://zenodo.org/records/3555552

Evaluation

Clean Error and Corruption Error(Mean)

CIFAR10 python cifar.py --resume [model_best.pth.tar path] --evaluate

CIFAR100 python cifar.py --resume [model_best.pth.tar path] --evaluate --dataset cifar100

===============================================================

About

Multi block + Random Erasing Data Augmentation. Experiments on CIFAR10, CIFAR100 and Fashion-MNIST

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%