.. _fwflag_fs_params:
=============
--fs_params
=============
Switch
======

--fs_params

Description
===========

Enables passing explicit feature-selection parameters via :doc:`fwflag_k_best`/:doc:`fwflag_pca_comp`, bundled into a ``{'kbest': ..., 'pca': ...}`` dict.

Argument and Default Value
==========================

No argument. Boolean switch, default: ``False``.

Details
=======

Verified: running with ``--fs_params --k_best 10 --pca_comp 5`` completed a full :doc:`fwflag_combo_test_regression` run without error. The resulting ``featureSelectionParameters`` dict is passed through to ``RegressionPredictor.testControlCombos()``, but its concrete effect on the feature-selection pipeline wasn't independently confirmed in the output (``{modelFS_desc}`` still showed ``None`` in this run) — it likely only changes behavior when combined with a real :doc:`fwflag_feature_selection` pipeline that consumes these values.

Other Switches
==============

Optional Switches:

* :doc:`fwflag_k_best`, :doc:`fwflag_pca_comp`
* :doc:`fwflag_feature_selection`

Example Commands
================

Using the packaged ``dla_tutorial`` dataset (see :doc:`../../tutorials/tut_dla`):

.. code-block:: bash

	dlatkInterface.py -d dla_tutorial -t msgs -c user_id -f 'feat$1to3gram$msgs$user_id$0_05' \
	--outcome_table blog_outcomes --outcomes age --combo_test_regression --fs_params --k_best 10 --pca_comp 5