DEV Community

Super Kai (Kazuya Ito)
Super Kai (Kazuya Ito)

Posted on • Edited on

logical_xor and logical_not in PyTorch

Buy Me a Coffee

*My post explains logical_and() and logical_or().

logical_xor() can do logical XOR with two of the 0D or more D tensors of zero or more elements, getting the 0D or more D tensor of zero or more boolean values as shown below:

*Memos:

  • logical_xor() can be used with torch or a tensor.
  • The 1st argument(input) with torch or using a tensor(Required-Type:tensor of int, float, complex or bool).
  • The 2nd argument with torch or the 1st argument with a tensor is other(Required-Type:tensor of int, float, complex or bool).
  • There is out argument with torch(Optional-Default:None-Type:tensor): *Memos:
    • out= must be used.
    • My post explains out argument.
  • Zero is True and nonzero is False.
import torch

tensor1 = torch.tensor(True)
tensor2 = torch.tensor([[True, False, True, False],
                        [False, True, False, True]])
torch.logical_xor(input=tensor1, other=tensor2)
tensor1.logical_xor(other=tensor2)
torch.logical_xor(input=tensor2, other=tensor1)
# tensor([[False, True, False, True],
#         [True, False, True, False]])

tensor1 = torch.tensor(False)
tensor2 = torch.tensor([[True, False, True, False],
                        [False, True, False, True]])
torch.logical_xor(input=tensor1, other=tensor2)
torch.logical_xor(input=tensor2, other=tensor1)
# tensor([[True, False, True, False],
#         [False, True, False, True]])

tensor1 = torch.tensor([True, False])
tensor2 = torch.tensor([[[True, False], [True, False]],
                        [[False, True], [False, True]]])
torch.logical_xor(input=tensor1, other=tensor2)
torch.logical_xor(input=tensor2, other=tensor1)
# tensor([[[False, False], [False, False]],
#         [[True, True], [True, True]]])

tensor1 = torch.tensor([7, 0])
tensor2 = torch.tensor([[[1, 0], [-1, 0]],
                        [[0, 2], [0, -2]]])
torch.logical_xor(input=tensor1, other=tensor2)
torch.logical_xor(input=tensor2, other=tensor1)
# tensor([[[False, False], [False, False]],
#         [[True, True], [True, True]]])

tensor1 = torch.tensor([7.3, 0.3])
tensor2 = torch.tensor([[[1.3, 0.3], [-1.0, 0.0]],
                        [[0.3, 2.3], [0.0, -2.0]]])
torch.logical_xor(input=tensor1, other=tensor2)
torch.logical_xor(input=tensor2, other=tensor1)
# tensor([[[False, False], [False, True]],
#         [[False, False], [True, False]]])

tensor1 = torch.tensor([7.+3.j, 0.+3.j])
tensor2 = torch.tensor([[[1.+3.j, 0.+3.j], [-1.+0.j, 0.+0.j]],
                        [[0.+3.j, 2.+3.j], [0.+0.j, -2.+0.j]]])
torch.logical_xor(input=tensor1, other=tensor2)
torch.logical_xor(input=tensor2, other=tensor1)
# tensor([[[False, False], [False, True]],
#         [[False, False], [True, False]]])
Enter fullscreen mode Exit fullscreen mode

logical_not() can do logical NOT with the 0D or more D tensor of zero or more elements, getting the 0D or more D tensor of zero or more boolean values as shown below:

*Memos:

  • logical_not() can be used with torch or a tensor.
  • The 1st argument(input) with torch or using a tensor(Required-Type:tensor of int, float, complex or bool).
  • There is out argument with torch(Optional-Default:None-Type:tensor): *Memos:
    • out= must be used.
    • My post explains out argument.
  • Zero is True and nonzero is False.
import torch

my_tensor = torch.tensor(True)

torch.logical_not(input=my_tensor)
my_tensor.logical_not()
# tensor(False)

my_tensor = torch.tensor([True, False, True, False])

torch.logical_not(input=my_tensor)
# tensor([False, True, False, True])

my_tensor = torch.tensor([[True, False, True, False],
                          [False, True, False, True]])
torch.logical_not(input=my_tensor)
# tensor([[False, True, False, True],
#         [True, False, True, False]])

my_tensor = torch.tensor([[[True, False], [True, False]],
                          [[False, True], [False, True]]])
torch.logical_not(input=my_tensor)
# tensor([[[False, True], [False, True]],
#         [[True, False], [True, False]]])

my_tensor = torch.tensor([[[1, 0], [-1, 0]],
                          [[0, 2], [0, -2]]])
torch.logical_not(input=my_tensor)
# tensor([[[False, True], [False, True]],
#         [[True, False], [True, False]]])

my_tensor = torch.tensor([[[1.3, 0.3], [-1.0, 0.0]],
                          [[0.3, 2.3], [0.0, -2.0]]])
torch.logical_not(input=my_tensor)
# tensor([[[False, False], [False, True]],
#         [[False, False], [ True, False]]])

my_tensor  = torch.tensor([[[1.+3.j, 0.+3.j], [-1.+0.j, 0.+0.j]],
                           [[0.+3.j, 2.+3.j], [0.+0.j, -2.+0.j]]])
torch.logical_not(input=my_tensor)
# tensor([[[False, False], [False, True]],
#         [[False, False], [True, False]]])
Enter fullscreen mode Exit fullscreen mode

Sentry image

See why 4M developers consider Sentry, “not bad.”

Fixing code doesn’t have to be the worst part of your day. Learn how Sentry can help.

Learn more

Top comments (0)

Billboard image

The Next Generation Developer Platform

Coherence is the first Platform-as-a-Service you can control. Unlike "black-box" platforms that are opinionated about the infra you can deploy, Coherence is powered by CNC, the open-source IaC framework, which offers limitless customization.

Learn more