source: orange/Orange/doc/widgets/Regression/Pade.htm @ 9671:a7b056375472

Revision 9671:a7b056375472, 3.5 KB checked in by anze <anze.staric@…>, 2 years ago (diff)

Moved orange to Orange (part 2)

Line 
1<html>
2<head>
3<title>Pade</title>
4<link rel=stylesheet href="../../../style.css" type="text/css" media=screen>
5<link rel=stylesheet href="style-print.css" type="text/css" media=print></link>
6</head>
7
8<body>
9
10<h1>Pade</h1>
11
12<img class="screenshot" src="../icons/Pade.png">
13<p>Replaces a continuous class with a derivative or a MQC by one or more continuous attributes.</p>
14
15<h2>Channels</h2>
16
17<h3>Inputs</h3>
18
19<DL class=attributes>
20<DT>Classified Examples (ExampleTableWithClass)</DT>
21<DD>Input data set.</DD>
22</dl>
23
24<h3>Outputs</h3>
25
26<DL class=attributes>
27<DT>Classified Examples (ExampleTableWithClass)</DT>
28<DD>Output data set.</DD>
29</dl>
30
31<h2>Description</h2>
32
33<p>This widget implements several techniques for assessing partial derivatives of the class variable for the given set of examples. The derivative is appended to the example table as a new class attribute. The widget can compute either quantitative derivative by a chosen continuous attribute or a qualitative derivative by one or more attributes.</p>
34
35<P>The widget is implemented to cache some data. After, for instance, computing the derivatives by <code>x</code>and <code>y</code> separately, the widget has already stored all the data to produce the derivatives by both in a moment.</P>
36
37<table>
38<tr><td valign="top">
39<img class="screenshot" src="Pade.png" align="left">
40</td>
41<td valign="top">
42<p>The <span class="option">Attributes</span> box lists all continuous attributes and lets the user select the attribute by which she wants to compute the qualitative derivative. The selection is important only when the widget actually outputs a qualitative derivative (this depends on other settings, described below). Buttons <span class="option">All</span> and <span class="option">None</span> select the entire list and nothing.</p>
43
44<P>Derivatives by more than one attribute are mathematically questionable, and computing by many attributes can be slow and messy. Methods that are based on triangulation will include all attributes in the triangulation, regardless of the selection, but then compute only the selected derivatives.</P>
45
46<P>Box <span class="option">Method</span> determines the used method and its settings. Available methods are <span class="option">First triangle</span>, <span class="option">Star Regression</span>, <span class="option">Univariate Star Regression</span> and <span class="option">Tube Regression</span>. First triangle is unsuitable for data with non-negligible noise. Star regression seems to perform rather poor; the quantitative derivatives it computes are even theoretically wrong. Univariate Star Regression will handle noise well, but also work well for very complex functions (like sin(x)sin(y) across several periods). Tube regression is very noise resistant, which can lead it to oversimplify the model, yet it is the only method that does not use the triangulation and is thus capable of handling discrete attributes, unknown values and large number of dimensions. It may be slow when the number of examples is very large. Detailed description of these methods can be found in Zabkar and Demsar's papers.</P>
47
48<P><span class="option">Ignore differences below</span> lets the user set a threshold for qualitative derivatives.</P>
49
50<p>The widget can also put some data in meta attributes: the <span class="option">Qualitative constraint</span>, as described above, <span class="option">Derivatives of selected attributes</span> and the <span class="option">Original class attribute</span>.</p>
51
52<p>The changes take effect and the widget start processing when <span class="option">Apply</span> is hit.</p>
53
54
55</body>
56</html>
Note: See TracBrowser for help on using the repository browser.