.. _auto_dataset_pca_inverse:

*DATASET PCA (INVERSE)*
=======================

Reconstructs the original curve columns from principal-component scores using the parameters from dataset_pca_fit: x = (scores . components) * std + mean. The input must contain the <pc_prefix>_PC1..n columns (e.g. predicted PCs from ML Predict). If the curve columns were normalized before PCA, follow this with dataset_normalize_inverse to recover fully raw values.

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

Tagged: ``pca``, ``inverse``, ``reconstruction``, ``dimensionality_reduction``, ``postprocessing``, ``prediction``, ``curves``.

Inputs
------

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

   * - Label
     - ID
     - Type
     - Default
     - Required
     - Description
   * - Dataset
     - dataset
     - dataset
     - —
     - ✓
     - Dataset containing the <pc_prefix>_PC1..n score columns to invert; array-of-row-objects.
   * - PCA Parameters
     - pca_parameters
     - json
     - —
     - ✓
     - PCA parameters object produced by dataset_pca_fit (components, mean, std, columns, pc_prefix).

Outputs
-------

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

   * - Label
     - ID
     - Type
     - Description
   * - Reconstructed Dataset
     - dataset
     - dataset
     - Input rows with the reconstructed curve columns (from params['columns']).
   * - Status
     - status
     - string
     - Summary of reconstructed columns.

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

- data.dataset.transform
- data.statistics

.. raw:: html

   <hr style="margin-top:2em">
   <p style="font-size:11px;color:#888">
   Auto-generated from <code>platform</code> schema. Worker id: <code>dataset_pca_inverse</code>. Schema hash: <code>2103cf42920f</code>. Hand-curated docs in <code>workerexamples/</code> override this page when present.
   </p>