source: orange/docs/widgets/rst/classify/logisticregression.rst @ 11404:1a7b773d7c7b

Revision 11404:1a7b773d7c7b, 2.5 KB checked in by Ales Erjavec <ales.erjavec@…>, 13 months ago (diff)

Replaced the use of :code: role with :obj:

RevLine 
[11050]1.. _Logistic Regression:
2
3Logistic Regression Learner
4===========================
5
6.. image:: ../icons/LogisticRegression.png
7
8Logistic Regression Learner
9
10Signals
11-------
12
13Inputs:
14
15
16   - Examples (ExampleTable)
17      A table with training examples
18
19
20Outputs:
21
22   - Learner
[11359]23      The logistic regression learning algorithm with settings as specified
24      in the dialog.
[11050]25
26   - Logistic Regression Classifier
27      Trained classifier (a subtype of Classifier)
28
29
[11404]30Signal :obj:`Logistic Regression Classifier` sends data only if the learning
31data (signal :obj:`Examples` is present.
[11050]32
33Description
34-----------
35
[11359]36This widget provides a graphical interface to the logistic regression
37classifier.
[11050]38
[11359]39As all widgets for classification, this widget provides a learner and
40classifier on the output. Learner is a learning algorithm with settings
41as specified by the user. It can be fed into widgets for testing learners,
42for instance :ref:`Test Learners`. Classifier is a logistic regression
43classifier (a subtype of a general classifier), built from the training
44examples on the input. If examples are not given, there is no classifier
45on the output.
[11050]46
[11359]47The widget requires - due to limitations of the learning algorithm - data with
48binary class.
[11050]49
50.. image:: images/LogisticRegression.png
51   :alt: Logistic Regression Widget
52
[11359]53Learner can be given a name under which it will appear in, say,
54:ref:`Test Learners`. The default name is "Logistic Regression".
[11050]55
[11359]56If :obj:`Stepwise attribute selection` is checked, the learner will
57iteratively add and remove the attributes, one at a time, based on their
58significance. The thresholds for addition and removal of the attribute are
59set in :obj:`Add threshold` and :obj:`Remove threshold`. It is also possible
60to limit the total number of attributes in the model.
[11050]61
[11359]62Independent of these settings, the learner will always remove singular
63attributes, for instance the constant attributes or those which can be
64expressed as a linear combination of other attributes.
[11050]65
[11359]66Logistic regression has no internal mechanism for dealing with missing
67values. These thus need to be imputed. The widget offers a number of options:
68it can impute the average value of the attribute, its minimum and maximum or
69train a model to predict the attribute's values based on values of other
70attributes. It can also remove the examples with missing values.
[11050]71
[11359]72Note that there also exist a separate widget for missing data imputation,
73:ref:`Impute`.
[11050]74
75
76Examples
77--------
78
[11359]79The widget is used just as any other widget for inducing classifier. See,
80for instance, the example for the :ref:`Naive Bayes`.
Note: See TracBrowser for help on using the repository browser.