mil-tokyo / bc_learning_image

Chainer implementation of between-class learning for Image classification https://arxiv.org/abs/1711.10284

BC learning for images

Implementation of Between-class Learning for Image Classification by Yuji Tokozume, Yoshitaka Ushiku, and Tatsuya Harada.

Our preliminary experimental results on CIFAR-10 and ImageNet-1K were already presented in ILSVRC2017 on July 26, 2017.

Between-class (BC) learning:

• We generate between-class examples by mixing two training examples belonging to different classes with a random ratio.
• We then input the mixed data to the model and train the model to output the mixing ratio.
• Original paper: Learning from Between-class Examples for Deep Sound Recognition by us (github)

Contents

• BC learning for images
  • BC: mix two images simply using internal divisions.
  • BC+: mix two images treating them as waveforms.
• Training of 11-layer CNN on CIFAR datasets

Setup

• Install Chainer v1.24 on a machine with CUDA GPU.
• Prepare CIFAR datasets.

Training

• Template:

    python main.py --dataset [cifar10 or cifar100] --netType convnet --data path/to/dataset/directory/ (--BC) (--plus)
  

• Recipes:

  • Standard learning on CIFAR-10 (around 6.1% error):

      python main.py --dataset cifar10 --netType convnet --data path/to/dataset/directory/
    

  • BC learning on CIFAR-10 (around 5.4% error):

      python main.py --dataset cifar10 --netType convnet --data path/to/dataset/directory/ --BC
    

  • BC+ learning on CIFAR-10 (around 5.2% error):

      python main.py --dataset cifar10 --netType convnet --data path/to/dataset/directory/ --BC --plus
    

• Notes:

  • It uses the same data augmentation scheme as fb.resnet.torch.
  • By default, it runs training 10 times. You can specify the number of trials by using --nTrials command.
  • Please check opts.py for other command line arguments.

Results

Error rate (average of 10 trials)

Learning	CIFAR-10	CIFAR-100
Standard	6.07	26.68
BC (ours)	5.40	24.28
BC+ (ours)	5.22	23.68

• Other results (please see paper):
  • The performance of Shake-Shake Regularization [1] on CIFAR-10 was improved from 2.86% to 2.26%.
  • The performance of ResNeXt [2] on ImageNet-1K was improved from 20.4% to 19.4% (single-crop top-1 validation error).

---

Reference

[1] X. Gastaldi. Shake-shake regularization. In ICLR Workshop, 2017.

[2] S. Xie, R. Girshick, P. Dollar, Z. Tu, and K. He. Aggregated residual transformations for deep neural networks. In CVPR, 2017.