What's this

Implementation of FractalNet by chainer

Dependencies

git clone https://github.com/nutszebra/fractal_net.git
cd fractal_net
git submodule init
git submodule update

How to run

python main.py -g 0

Details about my implementation

All hyperparameters and network architecture are the same as in [1] except for data-augmentation and learning rate schdedule.

• 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.

• Learning rate schedule
  Initial leranig rate is 0.06 and it is divided by 10 at [200, 300, 350, 375] epoch.

• Global drop path
  Implemented

• Local drop path
  Implemented

Cifar10 result

network	total accuracy (%)
FractalNet [1]	95.41
my implementation	93.77

References

FractalNet: Ultra-Deep Neural Networks without Residuals [1]