In this article, we are going to see how to compute the element-wise angle of a given input tensor in PyTorch.
torch.angle() method
Pytorch is an open-source deep learning framework available with a Python and C++ interface. Pytorch resides inside the torch module. In PyTorch, we will use torch.angle() method to compute the element-wise angle, the input is passed in the form of a tensor in radians. To compute the angle in the degree we multiply the angle in radian by 180/NumPy.pi, before moving further let’s see the syntax of the given method:
Syntax: torch.angle(tens, *, out=None)
Parameters:
- tens – This is our input tensor.
- out – This is our output tensor (optional).Â
Returns: Returns the elements-wise angle of given tensor.
Example 1:
In this example, we will compute element-wise angles in radians of a given 1-D tensor.
Python3
# Import the required libraries import torch   # define a complex tensor input = torch.tensor([1 - 2j, 2 + 1j, 3 - 2j, -4 + 3j, -5 - 2j])   # print the above define tensor print("\n Input Tensor: ", input)   # compute the elements-wise angle in radians ang = torch.angle(input)   # print the computed elements-wise angle in radians print("\n Elements-wise angles in radian: ", ang) |
Output:
Â
Example 2:
In this example, we will compute element-wise angles to convert the angle into a degree of the given 1-D tensor.
Python3
# Import the required libraries import torch from numpy import pi   # define a complex tensor input = torch.tensor([1 - 2j, 2 + 1j, 3 - 2j, -4 + 3j, -5 - 2j])   # print the above define tensor print("\n\nInput Tensor: ", input)   # compute the elements-wise angle in radians ang = torch.angle(input)   # convert the angle in degree deg = ang*180/pi   # print the computed elements-wise angle in degree print("\nElements-wise angles in degree: ", deg) |
Output:
Â
Example 3:
In this example, we will compute element-wise angles to convert the angle into a degree of the given 2-D tensor.
Python3
# Import the required libraries import torch from numpy import pi   # define a complex tensor input = torch.tensor([[1 - 2j, 2 + 3j, 3 - 3j],                       [4 + 3j, 5 - 4j, -6 + 2j],                       [-7 - 2j, 8 + 2j, 9 - 4j]])   # print the above define tensor print("\n Input Tensor:\n ", input)   # compute the elements-wise angle in radians radians = torch.angle(input)   # print the computed elements-wise angle in radians print("\n Elements-wise angles in radians:\n ", radians)   # convert the angle in degree degree = radians * 180/pi   # print the computed elements-wise angle in degree print("\n Elements-wise angles in degree:\n ", degree) |
Output:
Â
