kuanzi

Results 2 issues of kuanzi

It is ok for small scale models like MLP/LeNet5. However, when it comes to vgg16/ resnet18, it will always produce nan loss. The model structure configuration is below: `cfg =...

It is ok for small scale models like MLP/LeNet5. However, when it comes to vgg16/ resnet18, it will always produce nan loss. The model structure configuration is below: ` cfg...