Train CIFAR-10 in <7 seconds on an A100, the current world record.
Why do you think that https://github.com/karpathy/nanoGPT is a good alternative to hlb-CIFAR10
Train CIFAR-10 in <7 seconds on an A100, the current world record.
Why do you think that https://github.com/karpathy/nanoGPT is a good alternative to hlb-CIFAR10