About pre-trained models

[jimo17]

Hi, thank you for open sourcing. I am very interested in your thesis. Can you provide the 40-4 wide residual network pre-training model trained on CIFAR-10 and CIFAR-100 mentioned in the paper?