InverseGamma#
- class hybrid_aif.distributions.InverseGamma[source]#
Bases:
ModuleIndependent inverse-Gamma posteriors over per-dimension noise variances.
alpha/betaare(*parent_shape, d_obs): one (shape, rate) pair per observation dimension per parent configuration. Isotropic noise is thed_obs = 1case.- classmethod from_R_chol(R_chol, strength=1.0, anchor=False)[source]#
Prior centred on a known noise level:
point()reproducesR_cholexactly, andstrengthcounts pseudo-observations (α = strength/2).anchor=Truestores the initial parameters as the decay anchors.
- point()[source]#
R_cholunder the expected-precision substitution:diag(sqrt(β/α)).Not the mean covariance (
mean()): conditioning consumes a precision.
- stats(scatter, count)[source]#
Package one tick’s statistic: the responsibility-weighted residual scatter diagonal
(..., d_obs)and the responsibility mass(...,)(the degrees-of-freedom increment).
- __init__(alpha, beta, alpha0=None, beta0=None)#