So for hardware accelerated training with something like PyTorch, does anyone have a good comparison between Metal vs Cuda, both in terms of performance and cap

So for hardware accelerated training with something like PyTorch, does anyone have a good comparison between Metal vs Cuda, both in terms of performance and capabilities?