source: orange/docs/widgets/rst/regression/linear.rst @ 11359:8d54e79aa135

Revision 11359:8d54e79aa135, 1.4 KB checked in by Ales Erjavec <ales.erjavec@…>, 14 months ago (diff)

Cleanup of 'Widget catalog' documentation.

Fixed rst text formating, replaced dead hardcoded reference links (now using
:ref:), etc.

Line 
1.. _Linear Regression:
2
3Linear Regression Learner
4=========================
5
6.. image:: ../../../../Orange/OrangeWidgets/icons/Unknown.png
7    :alt: Linear Regression
8
9Learns a linear function of its input data.
10
11Channels
12--------
13
14Input
15    - Data (Table)
16        Input data table
17
18Output
19    - Learner
20        The learning algorithm with the supplied parameters
21
22    - Predictor
23        Trained regressor
24
25    - Model  Statisics
26        A data table containing trained model statistics
27
28
29Signal ``Predictor`` and ``Model Statistics`` send the output
30signal only if input signal ``Data`` is present.
31
32Description
33-----------
34
35Linear Regression widget construct a learner/predictor that learns a
36linear function from its input data. Furthermore `Lasso`_ and `Ridge`_
37regularization parameters can be specified.
38
39.. image:: images/LinearRegression.png
40    :alt: Linear Regression interface
41
42.. rst-class:: stamp-list
43
44    1. The learner/predictor name
45    2. Train an ordinary least squares or ridge regression model
46    3. If ``Ridge lambda`` is checked the learner will build a ridge regression
47       model with 4 as the ``lambda`` parameter.
48    4. Ridge lambda parameter.
49    5. Use `Lasso`_ regularization.
50    6. The Lasso bound (bound on the beta vector L1 norm)
51    7. Tolerance (any beta value lower then this will be forced to 0)
52
53.. _`Lasso`: http://en.wikipedia.org/wiki/Least_squares#LASSO_method
54
55.. _`Ridge`: http://en.wikipedia.org/wiki/Ridge_regression
Note: See TracBrowser for help on using the repository browser.