pystra.calibration.factors

Use it: User guide · Worked example · Theory

Target design solving and specialist design-point factor derivation.

The coefficient and matrix methods preserve the Caprani and Khan (Structural Safety, 2023) reference cases. They assume a separable resistance-scale design rule, including multiplicative resistance/load model errors. They are not a general decomposition of arbitrary nonlinear limit states.

Functions

derive_factors	Derive coefficient or matrix factors from fully converged target designs.
design_with_factors	Apply a selected factor set to the nominal design rule, by case/name.
select_factors	Select explicit extrema or retain per-case factors, recording governing cases.
solve_designs	Solve each case to a target and retain status and residuals.
verify_designs	Check a common design scale or an explicitly named set of designs.

Classes

CalibratedDesign	Target solve outcome; unsuccessful designs cannot enter factor derivation.
DesignValues	Factored design parameters in named case order.
DesignVerification	Achieved reliability and optional target margin for one supplied design.
FactorCalibrationProblem	Explicit inputs to the separable resistance-scale factor methods.
FactorSet	Immutable per-case factors with named axes and selection provenance.
GoverningFactor	Provenance for an extremal factor selection (ties retained).
TargetDesigns	Target designs and the input problem used to obtain them.