I'm currently using torch 1.12 for my thesis on neural A star algorithm and i'm not quite sure what meshgrid does. For example why x = torch.tensor([1, 2, 3]) y = torch.tensor([4, 5, 6]) torch.meshgrid(x, y) return tensor([[1, 1, 1], [2, 2, 2], [3, 3, 3]]) tensor([[4, 5, 6], [4, 5, 6], [4, 5, 6]]) and not tensor([[1, 1, 1], [2, 2, 2], [3, 3, 3]]) tensor([[4, 4, 4], [5, 5, 5], [6, 6, 6]]) Continue reading...