source: orange/orange/doc/widgets/Data/DataSampler.htm @ 9399:6bbe263e8bcf

Revision 9399:6bbe263e8bcf, 2.5 KB checked in by mitar, 2 years ago (diff)

Renaming widgets catalog.

Line 
1<html>
2<head>
3<title>Data Sampler</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>Data Sampler</h1>
11
12<img class="screenshot" src="../icons/DataSampler.png">
13<p>Selects a subset of data instances from the input data set.</p>
14
15<h2>Channels</h2>
16
17<h3>Inputs</h3>
18
19<DL class=attributes>
20<DT>Examples (ExampleTable)</DT>
21<DD>Attribute-valued data set.</DD>
22</dl>
23
24<h3>Outputs</h3>
25
26<DL class=attributes>
27<DT>Sample (ExampleTable)</DT>
28<DD>Attribute-valued data set as sampled from the input data.</DD>
29<DT>Remaining Examples (ExampleTable)</DT>
30<DD>Data instances from input data set that are not included in the sampled data.</DD>
31</dl>
32
33<h2>Description</h2>
34
35<p>Data Sampler supports provides support for several means of
36sampling of the data from the input channel and outputs the sampled
37data set and complementary data set (with instances from the input set
38that are not included in the sampled data set). Output is set when the
39input data set is set to the widget or after <span class="option">Sample Data</span> is
40pressed.</p>
41
42<img class="screenshot"
43src="DataSampler.png" alt="Data Sampler" border=0>
44
45<p>Sampling may be stratified: if input data contains a class,
46sampling will try to match its class distribution in the output data
47sets.</p>
48
49<p>Several types of sampling are supported. <span class="option">Random
50sampling</span> can draw a
51fixed number of instances or create a data set with a size set as
52a proportion of instances from the input data set. In repeated
53sampling, an data instance may be included in a sampled data several
54times (like in bootstrap).</p>
55
56<p><span class="option">Cross validation</span>,
57<span class="option">Leave-one-out</span> or sampling that creates
58<span class="option">Multiple subsets</span> of preset sample sizes relative to the input data set
59(like random sampling) all create several data samples. Which one is
60send to the output is determined by the data set index in <span class="option">Fold/Group</span>
61(indices start with 1).</p>
62
63<h2>Examples</h2>
64
65<p>Schema where we have sampled 10 data instances from Iris data set
66and presented this selection in <a
67href="Visualize/Scatterplot.htm">Scatterplot</a> widget is shown
68below.</p>
69
70<p><a href="DataSampler-Example.gif"><small>Click to enlarge</small<br/></small><img src="DataSampler-Example-S.gif"
71alt="Schema with Data Sampler" class="screenshot" border=0 /></a></p>
72
73</body>
74</html>
Note: See TracBrowser for help on using the repository browser.