pytorch-cifar10 icon indicating copy to clipboard operation
pytorch-cifar10 copied to clipboard

The state-of-the-art algorithms on CIFAR-10 dataset in PyTorch

Results 5 pytorch-cifar10 issues
Sort by recently updated
recently updated
newest added

Did you manage to obtain the same results as reported in the papers you referenced? E.g., 97.69% accuracy for ShakeDrop and such?

In 73th line of shakedrop/model.py: ` out = self.shake_shake(out, alpha, drop_factor)` I don't now why u use the shake_shake here?

I think some of the paper/post links in the readme might be dead

In the original paper, shakedrop is given as **G(x) = x + (bl + a + bl*a)F(x)**, but your application makes **out = (drop_factor + alpha + alpha * (1...

![cifar1](https://user-images.githubusercontent.com/16641054/46775076-8b17e480-cd40-11e8-9501-89c6fbca36bd.jpg)