Buffer protocol means PyTorch can have a torch.tensor method, that can create a tensor of anything implementing this protocol (which NumPy arrays do), and likew

Buffer protocol means PyTorch can have a torch.tensor method, that can create a tensor of anything implementing this protocol (which NumPy arrays do), and likewise numpy.array can be constructed out of PyTorch array because PyTorch Tensor implements this protocol. E.g. they don't need to know about each other (so no dependency), only about the protocol.