- class estimator.nd.NoiseDistribution(stddev: float, mean: float = 0, n: int | None = None, bounds: tuple = (None, None), density: float = 1.0, tag: str = '')#
All noise distributions are instances of this class.
Call self as a function.
Sample a_1, …, a_η, b_1, …, b_η and return Σ(a_i - b_i).
DiscreteGaussian(stddev[, mean, n])
A discrete Gaussian distribution with standard deviation
DiscreteGaussianAlpha(alpha, q[, mean, n])
A discrete Gaussian distribution with standard deviation α⋅q/√(2π) per component.
SparseTernary(n, p[, m])
Distribution of vectors of length
pentries of 1 and
mentries of -1, rest 0.
Uniform(a, b[, n])
Uniform distribution ∈
[a,b], endpoints inclusive.
q, with balanced representation.
__init__(stddev[, mean, n, bounds, density, tag])
Compute the size of the support covering the probability given as fraction.
We consider a distribution "sparse" if its density is < 1/2.