## DEV Community

Super Kai (Kazuya Ito)

Posted on • Updated on

# minimum() and maximum() in PyTorch

*Memos:

minimum() can get the 0D or more D tensor of zero or more minimum elements prioritizing `nan` from two of the 0D or more D tensors of zero or more elements as shown below:

*Memos:

• `minimum()` can be used with torch or a tensor.
• The 1st argument(`input`) with `torch` or using a tensor(Required-Type:`tensor` of `int`, `float` or `bool`).
• The 2nd argument with `torch` or the 1st argument is `other`(Required-Type:`tensor` of `int`, `float` or `bool`).
• There is `out` argument with `torch`(Optional-Type:`tensor`): *Memos:
• `out=` must be used.
• My post explains `out` argument.
• `nan` is taken if there are a number and `nan`.
``````import torch

tensor1 = torch.tensor([5., float('nan'), 4., float('nan')])
tensor2 = torch.tensor([[7., 8., float('nan'), float('nan')],
[-9., 2., 0., -6.]])
torch.minimum(input=tensor1, other=tensor2)
tensor1.minimum(other=tensor2)
# tensor([[5., nan, nan, nan],
#         [-9., nan, 0., nan]])

tensor1 = torch.tensor(5.)
tensor2 = torch.tensor([[[7., 8.], [float('nan'), float('nan')]],
[[-9., 2.], [0., -6.]]])
torch.minimum(input=tensor1, other=tensor2)
# tensor([[[5., 5.], [nan, nan]],
#         [[-9., 2.], [0., -6.]]])

tensor1 = torch.tensor(5)
tensor2 = torch.tensor([[[7, 8], [-5, -1]],
[[-9, 2], [0, -6]]])
torch.minimum(input=tensor1, other=tensor2)
# tensor([[[5, 5], [-5, -1]],
#         [[-9, 2], [0, -6]]])

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

maximum() can get the 0D or more D tensor of zero or more maximum elements prioritizing `nan` from two of the 0D or more D tensors of zero or more elements as shown below:

*Memos:

• `maximum()` can be used with `torch` or a tensor.
• The 1st argument(`input`) with `torch` or using a tensor(Required-Type:`tensor` of `int`, `float` or `bool`).
• The 2nd argument with `torch` or the 1st argument is `other`(Required-Type:`tensor` of `int`, `float` or `bool`).
• There is `out` argument with `torch`(Optional-Type:`tensor`): *Memos:
• `out=` must be used.
• My post explains `out` argument.
• `nan` is taken if there are a number and `nan`.
``````import torch

tensor1 = torch.tensor([5., float('nan'), 4., float('nan')])
tensor2 = torch.tensor([[7., 8., float('nan'), float('nan')],
[-9., 2., 0., -6.]])
torch.maximum(input=tensor1, other=tensor2)
tensor1.maximum(other=tensor2)
# tensor([[7., nan, nan, nan],
#         [5., nan, 4., nan]])

tensor1 = torch.tensor(5.)
tensor2 = torch.tensor([[[7., 8.], [float('nan'), float('nan')]],
[[-9., 2.], [0., -6.]]])
torch.maximum(input=tensor1, other=tensor2)
# tensor([[[7., 8.], [nan, nan]],
#         [[5., 5.], [5., 5.]]])

tensor1 = torch.tensor(5)
tensor2 = torch.tensor([[[7, 8], [-5, -1]],
[[-9, 2], [0, -6]]])
torch.maximum(input=tensor1, other=tensor2)
# tensor([[[7, 8], [5, 5]],
#         [[5, 5], [5, 5]]])

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