.. _auto_solverai_train:

*DEEP LEARNING SOLVER*
======================

Trains a deep learning model using the SOLVERAI engine on a supplied tabular dataset. Specify independent (input) and dependent (target) columns to define the supervised learning task; the worker outputs a serialised model checkpoint for downstream inference or deployment.

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

Tagged: ``checkpoint``, ``deep_learning``, ``neural_network``, ``simulation``, ``solverai``, ``supervised``, ``tabular``, ``training``.

Inputs
------

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

   * - Label
     - ID
     - Type
     - Default
     - Required
     - Description
   * - Input
     - input
     - dataset
     - —
     - ✓
     - Tabular training dataset (any format resolvable by the platform); must contain both the feature columns and target columns as named headers.
   * - Input Columns
     - input_cols
     - scalar
     - —
     - 
     - Comma-separated list of column names to use as independent (feature) variables; leave blank to let SOLVERAI infer them automatically from the dataset.
   * - Target Columns
     - target_cols
     - scalar
     - —
     - 
     - Comma-separated list of column names to use as dependent (target/label) variables; leave blank to let SOLVERAI infer them automatically from the dataset.

Outputs
-------

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

   * - Label
     - ID
     - Type
     - Description
   * - Checkpoint
     - checkpoint
     - text
     - Serialised model checkpoint path/reference (text) produced after training completes; pass this to a SOLVERAI inference worker to generate predictions on new data.

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

- ai_ml.supervised.classification
- ai_ml.supervised.regression
- ai_ml.surrogate

.. raw:: html

   <hr style="margin-top:2em">
   <p style="font-size:11px;color:#888">
   Auto-generated from <code>platform</code> schema. Worker id: <code>solverai_train</code>. Schema hash: <code>daf81092cee3</code>. Hand-curated docs in <code>workerexamples/</code> override this page when present.
   </p>