Changeset 11791:a8f3f1f6ab72 in orange


Ignore:
Timestamp:
12/05/13 08:09:54 (4 months ago)
Author:
blaz <blaz.zupan@…>
Branch:
default
Message:

Update of documentation for Select Attributes widget.

Files:
3 added
3 edited
1 moved

Legend:

Unmodified
Added
Removed
  • Orange/OrangeWidgets/Data/OWDataDomain.py

    r11784 r11791  
    55 
    66NAME = "Select Attributes" 
    7 DESCRIPTION = "Supports manual selection of data features and compostion." 
     7DESCRIPTION = "Manual selection of data features and composition of data domain." 
    88ICON = "icons/SelectAttributes.svg" 
    99PRIORITY = 1100 
  • docs/widgets/rst/data/selectattributes.rst

    r11778 r11791  
    88 
    99 
     10 
    1011Signals 
    1112------- 
    1213 
    1314Input: 
    14   - Examples (ExampleTable) 
     15  - Data 
    1516      Attribute-valued data set. 
    1617 
    1718Outputs: 
    18   - Examples (ExampleTable) 
     19  - Data 
    1920      Attribute-valued data set composed using the domain 
    20       specification constructed using the widget. 
     21      specification from the widget. 
    2122 
    2223Description 
     
    2425 
    2526Select Attributes widget is used to manually compose your data 
    26 domain, that is, to decide which attributes will be used and how. 
     27domain. User can decide which attributes will be used and how. 
    2728Orange distinguishes between ordinary attributes, an (optional) class 
    2829attributes and meta attributes. For instance, for building a classification 
     
    3536before the name of the attribute (D, C, S, respectively). 
    3637 
    37 Changes made to the domain are propagated to the output by pressing an 
    38 Apply button. Reset will present the attributes as defined in original 
    39 domain in the data set from the input signal 
     38.. image:: images/SelectAttributes-stamped.png 
     39   :alt: Select Attributes widget 
    4040 
    41 .. image:: images/SelectAttributes.png 
    42    :alt: Select Attributes widget example 
     411. Left-out data attributes from the input data file that will not be in the 
     42   data domain of the output data file. 
     43#. Data attributes in the new data file. 
     44#. A class attribute. If none, the new data set will be classless. 
     45#. Meta attributes of the new data file. The attributes are included in the 
     46   data set but are, for most of the methods, not considered in data analysis. 
     47#. Apply change of the data domain and send the new data file to the output 
     48   channel of the widget. 
     49#. Reset the domain composition to that of the input data file. 
     50#. Add an entry on domain data composition to the current report. 
    4351 
    44 Examples 
    45 -------- 
    4652 
    47 Below is a simple example how to use this widget. The input is fed directly 
    48 from  the :ref:`File` widget, and the output to the :ref:`Data Table` widget. 
    49 We have also linked the former to the File widget so that one can inspect the 
    50 difference in the domain composition. 
     53Example 
     54------- 
    5155 
    52 .. image:: images/SelectAttributes_schema.png 
     56In a workflow below an input data from the :ref:`File` widget 
     57is fed to the Select Attributes widget, which outputs the data to 
     58the :ref:`Data Table`. 
     59 
     60.. image:: images/SelectAttributes-Schema.png 
    5361   :alt: Select Attributes schema 
     62 
     63We used the schema to redefine the classification problem in 
     64heart disease data set. Originally, the task was to predict if the patient 
     65has a coronary artery diameter narrowing. We change the problem to that 
     66of gender classification based on age, chest pain and cholesterol level, 
     67and informatively kept the diameter narrowing attribute as a meta attribute. 
     68 
     69.. image:: images/SelectAttributes-Example.png 
     70   :alt: An example with Select Attributes widget 
Note: See TracChangeset for help on using the changeset viewer.