jaxvacua.freezer.LightSpectrum

class LightSpectrum(masses, eigenvalues, m2_min, stable, dw_residual, reduction, info)

Bases: object

Container for the light (reduced) mass spectrum returned by Freezer.light_mass_spectrum().

masses

Signed light masses \(\mathrm{sign}(m^2)\sqrt{|m^2|}\), sorted ascending.

Type:

np.ndarray

eigenvalues

Raw generalised eigenvalues \(\lambda = 2 m^2\) of \(H_{\rm eff}\,v = \lambda\,K_{\rm eff}\,v\).

Type:

np.ndarray

m2_min

Smallest light mass squared, \(0.5\,\min\lambda\).

Type:

float

stable

True if m2_min >= -rel_tol * max|m^2| (flat directions allowed, tachyons rejected). Note the tolerance is scaled by the largest mass: a tachyon smaller than rel_tol * max|m^2| is reported stable, so consult m2_min and info["m2_dynamic_range"] directly for a strongly hierarchical spectrum.

Type:

bool

dw_residual

On-shell screen value max|DW_x| of the full F-terms (heavy component included) at the evaluation point. When x_full is supplied this is the stored-vacuum residual; otherwise it is the analytic-reconstructed point’s residual (the heavy solve’s accuracy).

Type:

float

reduction

Reduction scheme used ("frozen" / "schur" / "autodiff" / "tangent").

Type:

str

info

Auxiliary diagnostics – the reduced-metric condition number cond_Keff, the reduced-spectrum dynamic range m2_dynamic_range (max|m^2| / min|m^2|, flat directions excluded; large values mean the lightest masses are precision-limited relative to the heaviest), the mode count n_modes, and – on an early return – a reason string.

Type:

Dict[str, Any]

__init__(masses, eigenvalues, m2_min, stable, dw_residual, reduction, info)

Methods

__init__(masses, eigenvalues, m2_min, ...)

Attributes

masses
eigenvalues
m2_min
stable
dw_residual
reduction
info