LogGrowthModel#
Logistic-growth model fitter with regularized least-squares objective.
Parameters and usage: LogGrowthModel.
One row per: fitted group (the
groupbycolumns)Written to: the frame
analyze()returns; when it is the pipeline’s model, a CLI run writesdeliverables/LogGrowthModel.csvand.parquetColumns: 14, each a pattern (see Placeholders)
Placeholders
Some column names depend on how you run the operation. They are shown as patterns, with these placeholders:
<metric>The fitted column without its metric-family prefix: fitting
Size_AreawritesLogGrowthModel_Area_r.
LOG_GROWTH_MODEL#
Column |
Description |
Type |
|---|---|---|
|
The intrinsic growth rate |
|
|
The carrying capacity |
|
|
The initial number of the colony size metric being fitted |
|
|
The regularization factor applied to the max specific growth rate and initial population size |
|
|
The penalty factor applied to relative difference of the carrying capacity from the largest measurement |
|
|
The growth rate of the colony calculated as (K*r)/4 |
|
|
The upper bound of the carrying capacity for model fitting |
MODEL_METRICS#
Every growth model writes these columns; see Shared columns.
Column |
Description |
Type |
|---|---|---|
|
The mean absolute error |
|
|
The mean squared error |
|
|
The root mean squared error |
|
|
The coefficient of determination |
|
|
The number of samples used for model fitting |
|
|
The loss of model fitting |
|
|
The output of the optimizer status |