PyTorch @ Operation Introduction: A Beginner Guide – PyTorch Tutorial

We may see @ operation in pytorch script. For example:

What does it mean? In this tutorial, we will introduce it for you.

@ in pytorch

In pytorch, @ operation is equal to torch.matmul().

The Difference Between torch.matmul() and torch.mul() – PyTorch Tutorial

For example:

import torch

x = torch.randn(3, 2)
y = torch.randn(2, 2)

z = x @ y
m = torch.matmul(x, y)

print("z=",z)
print("m=",m)

Run this code, we will see:

z= tensor([[ 0.5852,  0.6192],
        [ 0.9816,  1.6204],
        [-0.3757, -1.5175]])
m= tensor([[ 0.5852,  0.6192],
        [ 0.9816,  1.6204],
        [-0.3757, -1.5175]])

Here z and m are the same. It means:

@ = torch.matmul()