.. _measurement-op-linearcapandlagmodel: LinearCapAndLagModel ==================== Linear-softplus growth fitter with a softplus saturation ceiling. Parameters and usage: :py:class:`LinearCapAndLagModel `. - **One row per:** fitted group (the ``groupby`` columns) - **Written to:** the frame ``analyze()`` returns; when it is the pipeline's model, a CLI run writes ``deliverables/LinearCapAndLagModel.csv`` and ``.parquet`` - **Columns:** 14, each a pattern (see Placeholders) .. rubric:: Placeholders Some column names depend on how you run the operation. They are shown as patterns, with these placeholders: ```` The fitted column without its metric-family prefix: fitting ``Size_Area`` writes ``LinearCapAndLagModel_Area_v``. .. _measurement-info-linear-cap-and-lag-model: :doc:`LINEAR_CAP_AND_LAG_MODEL ` ----------------------------------------------------------------------------------------------- .. list-table:: Metric family: **LinearCapAndLagModel** :header-rows: 1 :class: phenotypic-measurement-table * - Column - Description - Type * - ``LinearCapAndLagModel__v`` - The post-lag phase growth rate. - :bdg-ref-success:`Tier 1 · Direct phenotype ` * - ``LinearCapAndLagModel__s0`` - The initial size - :bdg-ref-success:`Tier 1 · Direct phenotype ` * - ``LinearCapAndLagModel__lambda`` - The duration of the lag phase - :bdg-ref-success:`Tier 1 · Direct phenotype ` * - ``LinearCapAndLagModel__alpha`` - lag phase transition sharpness - :bdg-ref-primary:`Tier 2 · Descriptive trait ` * - ``LinearCapAndLagModel__smax`` - Carrying capacity used by the model. Either the user-provided scalar or the per-group observed maximum. - :bdg-ref-success:`Tier 1 · Direct phenotype ` * - ``LinearCapAndLagModel__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. - :bdg-ref-primary:`Tier 2 · Descriptive trait ` * - ``LinearCapAndLagModel__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). - :bdg-ref-secondary:`Quality ` :doc:`MODEL_METRICS ` ------------------------------------------------------------------------- Every growth model writes these columns; see :ref:`measurement-shared-columns`. .. list-table:: Metric family: **ModelMetrics** :header-rows: 1 :class: phenotypic-measurement-table * - Column - Description - Type * - ``ModelMetrics__MAE`` - The mean absolute error - :bdg-ref-secondary:`Quality ` * - ``ModelMetrics__MSE`` - The mean squared error - :bdg-ref-secondary:`Quality ` * - ``ModelMetrics__RMSE`` - The root mean squared error - :bdg-ref-secondary:`Quality ` * - ``ModelMetrics__R2`` - The coefficient of determination - :bdg-ref-secondary:`Quality ` * - ``ModelMetrics__NumSamples`` - The number of samples used for model fitting - :bdg-ref-secondary:`Quality ` * - ``ModelMetrics__OptimizerLoss`` - The loss of model fitting - :bdg-ref-secondary:`Quality ` * - ``ModelMetrics__OptimizerStatus`` - The output of the optimizer status - :bdg-ref-secondary:`Quality `