Training

When a stream is executed in IBM SPSS Modeler, the node from which the stream is executed can be thought of as bringing in the data from the previous node. This process continues all the way to the source node and repeats itself every time the stream is executed. Enabling a data cache can help you avoid the repetition of preprocessing by freezing the data at the cache, making it easier to reuse data for future analysis and improving speed and scalability.

To enable a data cache, select the node at which you want to place the cache from the Record or Field Ops palette. Then right click on the selected node, select Cache, and click Enable, as illustrated in Figure 1.

Figure 1: Selected node in a IBM SPSS Modeler stream

You have now set up a data cache. The next time you execute the stream, the cache will turn green as shown in Figure 2. This indicates that the data is frozen at the selected node and you do not need to run through all of the previous nodes.

Figure 2: Green indicates node with cached data