Implementation of Residual Networks (ResNet) by chainer
git clone https://github.com/nutszebra/residual_net.git
cd residual_net
git submodule init
git submodule update
python main.py -p ./ -g 0
All hyperparameters and network architecture are the same as in [1] except for data-augmentation.
- Data augmentation
Train: Pictures are randomly resized in the range of [32, 36], then 32x32 patches are extracted randomly and are normalized locally. Horizontal flipping is applied with 0.5 probability.
Test: Pictures are resized to 32x32, then they are normalized locally. Single image test is used to calculate total accuracy.
| network | depth | total accuracy (%) |
|---|---|---|
| ResNet [1] | 164 | 94.54 |
| my implementation | 164 | 94.39 |
| ResNet [1] | 1001 | 95.08 |
Identity Mappings in Deep Residual Networks [1]