Source code for phenotypic.schema._linear_cap_and_lag_model

"""Fitted parameters for the double-softplus growth model (with saturation ceiling)."""

from ._measurement_info import Entry
from ._tiers import DerivedMeasure




class LINEAR_CAP_AND_LAG_MODEL(DerivedMeasure):
    """Fitted parameters of the double-softplus growth model (with saturation).

    Output columns are **metric-qualified**: each header is
    ``LinearCapAndLagModel_<metric>_<parameter>``, where ``<metric>`` records
    the measurement the model was fit on (``self.on`` with its category prefix
    stripped, e.g. ``Shape_Area`` → ``Area``). For example, fitting on
    ``Shape_Area`` emits ``LinearCapAndLagModel_Area_v`` (post-lag growth rate)
    and ``LinearCapAndLagModel_Area_s0`` (initial size). The labels below are
    the ``<parameter>`` segment; the ``<metric>`` infix is filled in at fit
    time.
    """



    @classmethod
    def category(cls) -> str:
        return "LinearCapAndLagModel"




    @classmethod
    def header_scheme(cls) -> str:
        return "metric_qualified"


    v = Entry("v", "The post-lag phase growth rate.",
              tier=1, derivation_type="parameterization", derives_from="SIZE")
    s0 = Entry("s0", "The initial size",
               tier=1, derivation_type="parameterization", derives_from="SIZE")
    lam = Entry("lambda", "The duration of the lag phase",
                tier=1, derivation_type="parameterization", derives_from="SIZE")
    alpha = Entry("alpha", "lag phase transition sharpness",
                  tier=2, derivation_type="parameterization", derives_from="SIZE")
    smax = Entry(
        "smax",
        "Carrying capacity used by the model. Either the user-provided "
        "scalar or the per-group observed maximum.",
        tier=1, derivation_type="parameterization", derives_from="SIZE",
    )
    beta = Entry(
        "beta",
        "Saturation transition sharpness. Fitted per-group when a "
        "saturation shoulder is detected and ``beta`` is ``None`` at "
        "construction; held at the user-provided scalar (or the "
        "module default) when no shoulder is present.",
        tier=2, derivation_type="parameterization", derives_from="SIZE",
    )
    mode = Entry(
        "mode",
        "Fit variant selected per-group: 'fixed_beta' (beta held at "
        "the user-provided or module-default value) or 'fitted_beta' "
        "(beta fitted as a 5th free parameter when a saturation "
        "shoulder is detected).",
        derivation_type="diagnostic",
    )