43
Building Streams
A runnable node that processes stream data is known as a terminal node. A modeling or
output node is a terminal node if it is located at the end of a stream or stream branch. Youcannot
connect further nodes to a terminal node.
Note: You can customize the Nodes palette. Formo re information, see the topic Customizing
the Nodes Palette in Chapter 12 on p. 223.
Adding Nodes to a Stream
There are several ways to add nodes to a stream from the nodes palette:
Double-click a node on the palette. Note: Double-clicking a node automatically connects it
to the current stream. For more information, see the topic Co nnecting Nodes in a Stream
on p. 43.
Drag and drop a node from the palette to the stream canvas.
Click a node on the palette, and then click the stream canvas.
Select an appropriate option from the Insert menu of IBM® SPSS® Modeler.
Once you have added a node to the stream canvas, double-click the node to display its dialog box.
The available options depend on the type of node that you are adding. For informa tion about
specific controls within the dialog box, click its Help button.
Removing Nodes
Toremove a node from thedata stream, click it and ei ther press the Delete key,o r right-click and
select
Delete from the menu.
Connecting Nodes in a Stream
Nodes added to the stream canvas do not form a data stream unti l they have been connected.
Connections between the nodes indicate the direction of the data as it flows from one operation
to the next. There are a number of ways to connect nodes to form a stre am: double-clicking,
using the middle mouse button, or manually.
ToAdd and Connect Nodes by Double-Clicking
The simplest way to form a stream is to double-click nodes on the palette. This method
automatically connects the new node to the selected node on the stream canvas. For example, i f
the canvas contains a Database node, you can select this node and then double-click the next node
from the palette, such as a Derive node. This actio n automatically connects the Derive node to
the existing Database node. You can repeat this process until you have reac hed a terminal node,
such as a Histogram or Table node, at which point any new nodes will be connected to the last
non-terminal node upstream.