.asin()

The .asin() method in PyTorch computes the inverse sine (arcsine) of each element in a tensor, returning the angle in radians whose sine is the input value. The input values must be in the range [-1, 1], and the output values lie within [-π/2, π/2]. This method is useful in trigonometry, signal processing, and geometric transformations.

Syntax

torch.asin(input, *, out)

• input (Tensor): A tensor containing values in the range [-1, 1].
• out (Tensor, optional): A tensor to store the output. If provided, it must have the same shape as input.

The .asin() method returns a tensor where each element is the arcsine of the corresponding element in the input tensor.

Example

This example shows how to use the .asin() method to compute the inverse sine of tensor elements:

import torch

# Define a tensor with values in the range [-1, 1]
tensor = torch.tensor([0.0, 0.5, -1.0, 1.0])

# Compute inverse sine (arcsine)
asin_tensor = torch.asin(tensor)

print("Original Tensor:")
print(tensor)

print("\nArcsine Values:")
print(asin_tensor)

to clipboard
Copy to clipboard

This example results in the following output:

Original Tensor:
tensor([ 0.0000,  0.5000, -1.0000,  1.0000])

Arcsine Values:
tensor([ 0.0000,  0.5236, -1.5708,  1.5708])

to clipboard
Copy to clipboard

In this example, the output tensor shows the results of applying arcsin to each element:

• asin(0.0) = 0.0 → (since sin(0) = 0)
• asin(0.5) = π/6 ≈ 0.5236 → (since sin(π/6) = 0.5)
• asin(-1.0) = -π/2 ≈ -1.5708 → (since sin(-π/2) = -1)
• asin(1.0) = π/2 ≈ 1.5708 → (since sin(π/2) = 1)

Additional Notes

• Valid Range: Input values outside [-1, 1] will result in NaN (Not a number).
• In-Place Operation: Use .asin_() to modify the tensor directly.
• Complex Numbers: PyTorch’s .asin() supports complex tensors, computing the complex arcsine.