PyTorch .i0()

In PyTorch, the .i0() function (an alias of torch.special.i0()) computes the modified Bessel function of the first kind of order zero (I₀). This function is commonly used in signal processing, physics simulations, and statistical distributions such as the von Mises distribution for circular data.

Syntax

torch.i0(input, *, out=None)

Or, alternatively:

torch.special.i0(input, *, out=None)

Parameters:

• input (Tensor): The input tensor whose elements are the values at which to compute the modified Bessel function I₀.
• out (Tensor, optional): The output tensor to store the result.

Return value:

Returns a tensor containing the computed modified Bessel function of the first kind, order zero (I₀), for each element of the input tensor.

Example

In this example, torch.i0() calculates the modified Bessel function I₀ for each element in a 1D tensor:

import torch

# Create input tensor
x = torch.tensor([0.0, 1.0, 2.0, 3.0, 4.0])

# Compute the modified Bessel function
result = torch.i0(x)

print("Input values:", x)
print("I₀(x) values:", result)

Copy to clipboard
Copy to clipboard

The above code produces the following output:

Input values: tensor([0., 1., 2., 3., 4.])
I₀(x) values: tensor([1.0000, 1.2661, 2.2796, 4.8808, 11.3019])

Copy to clipboard
Copy to clipboard