.. _auto_kriging_interpolation_worker:

*KRIGING INTERPOLATION*
=======================

Fits a Kriging (Gaussian Process) surrogate model to a training dataset and uses it to interpolate predictions over a new set of input points. Also identifies the optimum (minimum or maximum) of the response surface. Use this worker when you need a smooth, probabilistic surrogate for design-space exploration or optimum-seeking tasks.

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

Tagged: ``design exploration``, ``gaussian process``, ``interpolation``, ``kriging``, ``optimization``, ``response surface``, ``surrogate``.

Inputs
------

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

   * - Label
     - ID
     - Type
     - Default
     - Required
     - Description
   * - Dataset
     - dataset
     - dataset
     - —
     - ✓
     - Training dataset containing independent variable columns and at least one target/response column; must be a structured tabular dataset with no missing values.
   * - Independents
     - independents
     - select
     - —
     - ✓
     - Column name(s) from the training dataset to treat as independent (input) variables for the Kriging model.
   * - Targets
     - targets
     - select
     - —
     - ✓
     - Column name(s) from the training dataset to treat as response (output) variables that the Kriging model will learn to predict.
   * - Objective
     - objective
     - select
     - min
     - ✓
     - Optimization direction for identifying the optimum on the fitted response surface; choose 'min' to minimize or 'max' to maximize the target — defaults to 'min'.
   * - Predict For
     - predict_dataset
     - dataset
     - —
     - 
     - Optional dataset of new input points (must share the same column names as the selected independents) for which the fitted Kriging model will generate predictions; leave empty to skip batch prediction.

Outputs
-------

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

   * - Label
     - ID
     - Type
     - Description
   * - Input Dataset
     - dataset
     - dataset
     - The original training dataset passed through, augmented with Kriging fitted values for diagnostic or downstream use.
   * - Predictions
     - predictions
     - dataset
     - Tabular dataset of Kriging-predicted response values for each row in the 'Predict For' input dataset; columns correspond to the selected target variables.
   * - Optimum
     - optimum
     - dataset
     - Single-row dataset identifying the input-variable combination and predicted response value that achieves the requested optimum (minimum or maximum) on the surrogate surface.

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

- ai_ml.surrogate
- 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>kriging_interpolation_worker</code>. Schema hash: <code>a8f2cffaf49f</code>. Hand-curated docs in <code>workerexamples/</code> override this page when present.
   </p>