In this article, we will see how to compute the logistic sigmoid function of Tensor Elements in PyTorch.
The torch.special.expit() & torch.sigmoid() methods are logistic functions in a tensor. torch.sigmoid() is an alias of torch.special.expit() method. So, these methods will take the torch tensor as input and compute the logistic function element-wise of the tensor.
Syntax:
torch.special.expit(tensor)
torch.sigmoid(tensor)Parameter:
- tensor is the input tensor
Return: Return the logistic function of elements with new tensor.
Example 1:
In this example, we are creating a one-dimensional tensor with 6 elements and returning the logistic sigmoid function of elements using the sigmoid() method.
Python3
import torch # create 1D tensor with 6 elements t1 = torch.arange(1, 13) # display print(t1) # Compute the logistic sigmoid # function of elements in the # above tensor print(torch.sigmoid(t1)) |
Output:
tensor([ 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12])
tensor([0.7311, 0.8808, 0.9526, 0.9820, 0.9933, 0.9975, 0.9991, 0.9997, 0.9999,
1.0000, 1.0000, 1.0000])
Example 2:
In this example, we are creating a one-dimensional tensor with 5 elements and returning the logistic sigmoid function of elements using torch.special.expit() method.
Python3
import torch # create 1D tensor with 5 elements t1 = torch.arange(1, 6) # display print(t1) # Compute the logistic sigmoid # function of elements in the # above tensor print(torch.special.expit(t1)) |
Output:
tensor([1, 2, 3, 4, 5])
tensor([0.7311, 0.8808, 0.9526, 0.9820, 0.9933])
Example 3:
In this example, we are creating a two-dimensional tensor with 3×3 elements, and returning the logistic sigmoid function of elements using sigmoid() method.
Python3
import torch # create 2D tensor with 3 elements each t1 = torch.tensor([[-20, 34, 56], [6, -9, 8]]) # display print(t1) # Compute the logistic sigmoid function # of elements in the above tensor print(torch.sigmoid(t1)) |
Output:
tensor([[-20, 34, 56],
[ 6, -9, 8]])
tensor([[2.0612e-09, 1.0000e+00, 1.0000e+00],
[9.9753e-01, 1.2339e-04, 9.9966e-01]])
Example 4:
In this example, we are creating a two-dimensional tensor with 3×3 elements each and, returning the logistic sigmoid function of elements using torch.special.expit() method.
Python3
import torch # create 2D tensor with 3 elements each t1 = torch.tensor([[-20, 34, 56, ], [78, 90, 8]]) # display print(t1) # Compute the logistic sigmoid # function of elements in the # above tensor print(torch.special.expit(t1)) |
Output:
tensor([[-20, 34, 56],
[ 6, -9, 8]])
tensor([[2.0612e-09, 1.0000e+00, 1.0000e+00],
[9.9753e-01, 1.2339e-04, 9.9966e-01]])
