spyrit.core.noise.Poisson

class spyrit.core.noise.Poisson(alpha: float = 1.0)

Bases: Module

Simulate measurements corrupted by Poisson noise

\[y \sim \mathcal{P}\left(\alpha z\right), \quad \text{with }z\ge 0\]

where \(\mathcal{P}\) is the Poisson distribution and \(\alpha\) represents the intensity of the noiseless measurements \(z\).

The class is constructed from the intensity \(\alpha\).

Args:

alpha (float): The intensity of the measurements. Defaults to 1.

Attributes:

alpha (float): Intensity of the measurements.

Example:

>>> noise = Poisson(10.0)
>>> z = torch.tensor([1, 3, 6])
>>> y = noise(z)
>>> print(y)
tensor([...])

Methods

forward(z)

Corrupt measurement by Poisson noise