source: orange/docs/widgets/rst/regression/earth.rst @ 10404:ce517a5eaa1e

Revision 10404:ce517a5eaa1e, 2.1 KB checked in by Ales Erjavec <ales.erjavec@…>, 2 years ago (diff)

Changed icon location.

Line 
1.. _Earth:
2
3Earth Learner
4=============
5
6.. image:: ../../../../Orange/OrangeWidgets/icons/Unknown.png
7   :alt: Earth Learner
8   
9Channels
10--------
11
12Inputs:
13   - Data (Table)
14
15Outputs:
16   - Learner
17        The Earth learning algorithm with parameters as specified in the dialog.
18       
19   - Predictor
20        Trained regressor
21       
22Signal ``Predictor`` sends the regressor only if signal ``Data`` is present. 
23
24Description
25-----------
26
27This widget constructs a Earth learning algorithm (an implementation of
28the `MARS - Multivariate Adaptive Regression Splines`_). As all widgets
29for classification and regression, this widget provides a learner and
30classifier/regressor on the output. Learner is a learning algorithm with
31settings as specified by the user. It can be fed into widgets for testing
32learners, for instance Test Learners.
33
34.. _`MARS - Multivariate Adaptive Regression Splines`: http://en.wikipedia.org/wiki/Multivariate_adaptive_regression_splines
35
36.. image:: images/Earth.png
37
38
39.. rst-class:: stamp-list
40
41    1. Learner/Predictor can be given a name under which it will appear
42       in other widgets (say ``Test Learners`` or ``Predictions``).
43   
44    2. The ``Max. term degree`` parameter specifies the degree of the
45       terms induced in the forward pass. For instance, if set to ``1`` 
46       the resulting model will contain only linear terms.
47       
48    3. The ``Max. terms`` specifies how many terms can be induces in the
49       forward pass. A special value ``Automatic`` instructs the learner
50       to set the limit automatically based on the dimensionality of the
51       data (``min(200, max(20, 2 * NumberOfAttributes)) + 1``)
52       
53    4. The ``Knot penalty`` is used in the pruning pass (hinge function 
54       penalty for the GCV calculation)
55
56
57After changing one or more settings, you need to push |5| ``Apply``,
58which will put the new learner on the output and, if the training
59examples are given, construct a new predictor and output it as well.
60
61
62Examples
63--------
64
65Lets use the learner to train a model on a data subset and test it on
66unseen instances.
67
68.. image:: images/Earth-Schema.png
Note: See TracBrowser for help on using the repository browser.