DEV Community

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

Posted on

eq and ne in PyTorch

Buy Me a Coffee

*Memos:

eq() can check if the zero or more elements of the 1st 0D or more D tensor are equal to the zero or more elements of the 2nd 0D or more D tensor element-wise, getting the 0D or more D tensor of zero or more elements as shown below:

*Memos:

  • eq() 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 or scalar 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.
  • The result is the higher D tensor which has more elements.
import torch

tensor1 = torch.tensor([5, 0, 3])
tensor2 = torch.tensor([7, 0, 3])

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

tensor1 = torch.tensor(5)
tensor2 = torch.tensor([[3, 5, 4],
                        [6, 3, 5]])
torch.eq(input=tensor1, other=tensor2)
torch.eq(input=tensor2, other=tensor1)
# tensor([[False, True, False],
#         [False, False, True]])

torch.eq(input=tensor1, other=3)
# tensor(False)

torch.eq(input=tensor2, other=3)
# tensor([[True, False, False],
#         [False, True, False]])

tensor1 = torch.tensor([5, 0, 3])
tensor2 = torch.tensor([[5, 5, 5],
                        [0, 0, 0],
                        [3, 3, 3]])
torch.eq(input=tensor1, other=tensor2)
torch.eq(input=tensor2, other=tensor1)
# tensor([[True, False, False],
#         [False, True, False], 
#         [False, False, True]])

torch.eq(input=tensor1, other=3)
# tensor([False, False, True])

torch.eq(input=tensor2, other=3)
# tensor([[False, False, False],
#         [False, False, False],
#         [True, True, True]])

tensor1 = torch.tensor([5., 0., 3.])
tensor2 = torch.tensor([[5., 5., 5.],
                        [0., 0., 0.],
                        [3., 3., 3.]])
torch.eq(input=tensor1, other=tensor2)
# tensor([[True, False, False],
#         [False, True, False], 
#         [False, False, True]])

torch.eq(input=tensor1, other=3.)
# tensor([False, False, True])

tensor1 = torch.tensor([5.+0.j, 0.+0.j, 3.+0.j])
tensor2 = torch.tensor([[5.+0.j, 5.+0.j, 5.+0.j],
                        [0.+0.j, 0.+0.j, 0.+0.j],
                        [3.+0.j, 3.+0.j, 3.+.0j]])
torch.eq(input=tensor1, other=tensor2)
# tensor([[True, False, False],
#         [False, True, False],
#         [False, False, True]])

torch.eq(input=tensor1, other=3.+0.j)
# tensor([False, False, True])

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

torch.eq(input=tensor1, other=True)
# tensor([True, False, True])
Enter fullscreen mode Exit fullscreen mode

ne() can check if the zero or more elements of the 1st 0D or more D tensor are not equal to the zero or more elements of the 2nd 0D or more D tensor element-wise, getting the 0D or more D tensor of zero or more elements as shown below:

*Memos:

  • ne() 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 or scalar 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.
  • not_equal() is the alias of ne().
import torch

tensor1 = torch.tensor([5, 0, 3])
tensor2 = torch.tensor([7, 0, 3])

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

tensor1 = torch.tensor(5)
tensor2 = torch.tensor([[3, 5, 4],
                        [6, 3, 5]])
torch.ne(input=tensor1, other=tensor2)
torch.ne(input=tensor2, other=tensor1)
# tensor([[True, False, True],
#         [True, True, False]])

torch.ne(input=tensor1, other=3)
# tensor(True)

torch.ne(input=tensor2, other=3)
# tensor([[False, True, True],
#         [True, False, True]])

tensor1 = torch.tensor([5, 0, 3])
tensor2 = torch.tensor([[5, 5, 5],
                        [0, 0, 0],
                        [3, 3, 3]])
torch.ne(input=tensor1, other=tensor2)
torch.ne(input=tensor2, other=tensor1)
# tensor([[False, True, True],
#         [True, False, True],
#         [True, True, False]])

torch.ne(input=tensor1, other=3)
# tensor([True, True, False])

torch.ne(input=tensor2, other=3)
# tensor([[True, True, True],
#         [True, True, True],
#         [False, False, False]])

tensor1 = torch.tensor([5., 0., 3.])
tensor2 = torch.tensor([[5., 5., 5.],
                        [0., 0., 0.],
                        [3., 3., 3.]])
torch.ne(input=tensor1, other=tensor2)
# tensor([[False, True, True],
#         [True, False, True],
#         [True, True, False]])

torch.ne(input=tensor1, other=3.)
# tensor([True, True, False])

tensor1 = torch.tensor([5.+0.j, 0.+0.j, 3.+0.j])
tensor2 = torch.tensor([[5.+0.j, 5.+0.j, 5.+0.j],
                        [0.+0.j, 0.+0.j, 0.+0.j],
                        [3.+0.j, 3.+0.j, 3.+.0j]])
torch.ne(input=tensor1, other=tensor2)
# tensor([[False, True, True],
#         [True, False, True],
#         [True, True, False]])

torch.ne(input=tensor1, other=3.+0.j)
# tensor([True, True, False])

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

torch.ne(input=tensor1, other=True)
# tensor([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)

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

👋 Kindness is contagious

Please leave a ❤️ or a friendly comment on this post if you found it helpful!

Okay