NoiseModel#

class deepinv.physics.NoiseModel(noise_model: Callable | None = None, rng: Generator | None = None)[source]#

Base class for noise model. NoiseModel can be combined via deepinv.physics.noise.NoiseModel.__mul__(),

Parameters:: rng (torch.Generator (Optional)) – a pseudorandom random number generator for the parameter generation. If is provided, it should be on the same device as the input.

__mul__(other)[source]#

Concatenates two noise \(N = N_1 \circ N_2\) via the mul operation

The resulting operator will add the noise from both noise models and keep the rng of \(N_1\).

Parameters:: other (deepinv.physics.noise.NoiseModel) – Physics operator \(A_2\)
Returns:: (deepinv.physics.noise.NoiseModel) concatenated operator

forward(input: Tensor, seed: int | None = None) → Tensor[source]#: Add noise to the input :param torch.Tensor input: input tensor :param int seed: the seed for the random number generator.

rand_like(input: Tensor, seed: int | None = None)[source]#: Equivalent to torch.rand_like but supports a pseudorandom number generator argument. :param int seed: the seed for the random number generator, if rng is provided.

randn_like(input: Tensor, seed: int | None = None)[source]#: Equivalent to torch.randn_like but supports a pseudorandom number generator argument. :param int seed: the seed for the random number generator, if rng is provided.

reset_rng()[source]#: Reset the random number generator to its initial state.

rng_manual_seed(seed: int | None = None)[source]#

Sets the seed for the random number generator.

Parameters:: seed (int) – the seed to set for the random number generator. If not provided, the current state of the random number generator is used. Note: it will be ignored if the random number generator is not initialized.