.. _auto_genetic_mutator:

*GENETIC MUTATOR*
=================

Runs a genetic-algorithm mutation loop over a dataset to find input configurations that minimize or maximize a target objective. Given a population of candidate designs, it iterates through mutation steps and returns both the evolved population and the best-found optimum point(s).

When to use
-----------

Tagged: ``doe``, ``evolutionary``, ``genetic``, ``genetic algorithm``, ``genetic mutator``, ``mutation``, ``mutator``, ``optimization``.

Inputs
------

.. list-table::
   :header-rows: 1
   :widths: 20 20 20 20 20 20

   * - Label
     - ID
     - Type
     - Default
     - Required
     - Description
   * - Dataset
     - dataset
     - dataset
     - —
     - 
     - Input dataset containing the design space; each row is a candidate design point (tabular format, columns must include the chosen independent and target variables).
   * - Independents
     - independents
     - select
     - —
     - ✓
     - Column name(s) from the dataset representing the independent/input variables (design parameters) to be mutated by the genetic algorithm.
   * - Targets
     - targets
     - select
     - —
     - ✓
     - Column name(s) from the dataset representing the target/response variable(s) that the objective function will minimize or maximize.
   * - Objective
     - objective
     - select
     - min
     - ✓
     - Optimization direction: 'min' to minimize the target or 'max' to maximize it; defaults to 'min'.
   * - Population Size
     - population_size
     - text
     - 30
     - ✓
     - Number of candidate individuals maintained in the population at each generation; larger values improve coverage but increase compute cost (default: 30).
   * - Number Of Iterations
     - num_iterations
     - text
     - 100
     - ✓
     - Total number of mutation/generation cycles the algorithm will execute before stopping; increase for more thorough search (default: 100).
   * - Mutation Probability
     - mut_prob
     - text
     - 0.2
     - ✓
     - Per-gene mutation probability in the range [0, 1]; controls how often individual design parameters are randomly perturbed each generation (default: 0.2).
   * - NOPTR
     - num_points_to_return
     - text
     - 1
     - ✓
     - Number of top-ranked (optimum) design points to return in the output optimum dataset; set to 1 to return only the single best solution (default: 1).

Outputs
-------

.. list-table::
   :header-rows: 1
   :widths: 20 20 20 20

   * - Label
     - ID
     - Type
     - Description
   * - Input Dataset
     - new_population
     - dataset
     - Dataset containing the full final-generation population of candidate design points after all mutation iterations, preserving independent variable columns.
   * - Optimum
     - optimum
     - dataset
     - Dataset containing the top N best-performing design points (N = num_points_to_return) ranked by the chosen objective, with their independent variable values and predicted target responses.

Disciplines
-----------

- ai_ml.preprocessing
- design_exploration.doe
- design_exploration.optimization

.. raw:: html

   <hr style="margin-top:2em">
   <p style="font-size:11px;color:#888">
   Auto-generated from <code>platform</code> schema. Worker id: <code>genetic_mutator</code>. Schema hash: <code>ad8fa705888b</code>. Hand-curated docs in <code>workerexamples/</code> override this page when present.
   </p>