spyrit.core.noise.PoissonApproxGaussSameNoise

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

Bases: Poisson

Gaussian approximation of Poisson noise

\[y \sim \alpha z + \sqrt{\alpha z} \cdot \mathcal{N}(0, 1), \quad \text{with }z\ge 0\]

where \(\alpha\) represents the intensity of the noiseless measurements \(z\), and \(\mathcal{N}(0, 1)\) is a Gaussian distribution with zero mean and unit variance.

This is an approximation of \(y \sim \mathcal{P}\left(\alpha z\right)\), where \(\mathcal{P}\) is the Poisson distribution. Computing the Gaussian approximation is faster than the original Poisson model.

Important

Contrary to PoissonApproxGauss, different noise realisations apply only to the last dimension of the input tensor. The same noise realisations are repeated to the first dimensions of the input tensor.

Args:

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

Attributes:

alpha (float): Intensity of the measurements.

Methods

forward(z)

Corrupt measurement by Gaussian approximation of Poisson noise