PyTorch .asinh()

The PyTorch method .asinh() returns the inverse hyperbolic sine of each element in a given tensor. It is useful for mathematical and scientific computations involving hyperbolic functions.

Syntax

torch.asinh(input, *, out=None) → Tensor

Copy to clipboard
Copy to clipboard

Parameters

• input: A tensor containing real or complex values of any dimension.
• out (Optional): The output tensor to store the result. If not specified, a new tensor is returned.

Returns

A new tensor containing the inverse hyperbolic sine of each element in the input tensor.

Example

The following example demonstrates how to use .asinh() to compute the inverse hyperbolic sine for the real-valued 1D tensor:

import torch

input_tensor = torch.tensor([0.125, 0.25, 0.5, 0.75])

output_tensor = torch.asinh(input_tensor)

print("Input Tensor:")
print(input_tensor)

print("\nOutput Tensor:")
print(output_tensor)

Copy to clipboard
Copy to clipboard

This example outputs:

Input Tensor:
tensor([0.1250, 0.2500, 0.5000, 0.7500])

Output Tensor:
tensor([0.1253, 0.2526, 0.5211, 0.8223])

Copy to clipboard
Copy to clipboard

The values in output_tensor represents the inverse hyperbolic sine of the corresponding values in input_tensor.