.. _fwflag_probability_csv:
===================
--probability_csv
===================
Switch
======

--probability_csv

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

Writes predicted class probabilities (rather than hard class labels) to a separate csv, for classification predictions. Aliased as ``--prob_csv``.

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

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

Details
=======

Used together with :doc:`fwflag_prediction_csv`: ``savePredictions = (args.pred_csv | args.prob_csv)`` gates whether predictions are saved at all, and the two together decide whether hard labels, probabilities, or both get written.

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

Required Switches:

* One of the classification prediction switches, e.g. :doc:`fwflag_combo_test_classifiers`, :doc:`fwflag_predict_classifiers`

Optional Switches:

* :doc:`fwflag_prediction_csv`

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

.. code-block:: bash

	dlatkInterface.py -d dla_tutorial -t msgs -c user_id -f 'feat$1gram$msgs$user_id$16to16' \
	--outcome_table blog_outcomes --outcomes is_student --combo_test_classifiers --probability_csv --output_name ./out