source: orange/docs/widgets/rst/data/file.rst @ 11050:e3c4699ca155

Revision 11050:e3c4699ca155, 2.7 KB checked in by Miha Stajdohar <miha.stajdohar@…>, 16 months ago (diff)

Widget docs From HTML to Sphinx.

Line 
1.. _File:
2
3File
4====
5
6.. image:: ../../../../Orange/OrangeWidgets/Data/icons/File_48.png
7   :alt: File widget icon
8   
9Signals
10-------
11
12Inputs:
13   - None
14
15Outputs:
16   - Examples (ExampleTable)
17        Attribute-valued data set read from the input file.
18
19Description
20-----------
21
22.. image:: images/File.png
23   :alt: File widget with loaded ionosphere.tab data set
24   
25This is the widget you will probably use in every schema. It reads the input data file
26(data table with examples) and sends the data set to the output channel. It maintains
27a history of most recently used data files. For convenience, the history also includes
28a directory with the sample data sets that come with Orange.
29
30To browse for a data file press .... Reload button reloads the file.
31
32The widget can read data from simple tab-delimited or comma-separated files, as well as
33files in C4.5 format and Weka's .arrf files.
34
35The info box at the bottom of the widget displays information on the number of examples,
36attributes and meta attributes, and information on class variable (if defined in a file).
37
38The controls in between are for more advanced users.
39
40Tab-delimited file can have user defined symbols for undefined values. The symbols for
41"don't care" and "don't know" values can be specified in the corresponding edit lines.
42The default values for "don't know" and "don't care" depend upon format. Most users will
43use tab-delimited files: keep the field empty or put a question mark in there and that's
44it. Most algorithms do not differ between don't know and don't care values, so consider
45them both to mean undefined.
46
47The part which most will probably neither understand nor need is related to construction
48of new attributes. It becomes important when multiple data sets are loaded, usually with
49multiple File widgets. A typical example is loading separate training and testing data
50from two files. In this case Orange will usually treat the attributes with the same name
51but appearing in different files as the same attribute, so a classifier which uses the
52attribute "petal length" from the first will use the attribute of the same name from
53the second. In cases when attributes from different files just accidentally bear different
54names, one can instruct Orange to either always construct new attribute or construct them when
55they differ in their domains. Use this option with great care (if at all).
56
57Examples
58--------
59
60Any of your schemas should probably start with the :ref:`File` widget. In the schema below,
61the widget is used to read the data that is then sent to both :ref:`Data Table` widget and
62to widget that displays :ref:`Attribute Statistics`.
63
64.. image:: images/File_schema.png
65   :alt: Example schema with File widget
Note: See TracBrowser for help on using the repository browser.