Importing an SPSS Modeler stream
You can import a stream (.str) that was created in SPSS Modeler Subscription or SPSS Modeler client.
- From your project, under Modeler flows, click New modeler flow.
- Select From File, select the .str file you want to import, and click Create.
If the imported stream contains one or more source (import) or export nodes, you'll be prompted to convert the nodes. Cloud Pak for Data will walk you through the migration process.
Watch the following video for an example of this easy process:
This video provides a visual method as an alternative to following the written steps in this documentation.
If the stream contains multiple import nodes that use the same data file, then you must first add that file to your project as a data asset before migrating because the conversion can't upload the same file to more than one import node. After adding the data asset to your project, reopen the flow and proceed with the migration using the new data asset. Nodes with the same name will be automatically mapped to project assets.
Configure export nodes to export to your project or to a connection. The following export nodes are supported:
- When migrating export nodes, you're converting node types that don't exist in Cloud Pak for Data. The nodes are converted to Data Asset export nodes or a connection. Due to a current limitation for automatically migrating nodes, only existing project assets or connections can be selected as export targets. These assets will be overwritten during export when the flow runs.
- To preserve any type or filter information, when an import node is replaced with Data Asset nodes, they're converted to a SuperNode.
- After migration, you can go back later and use the Convert button if you want to migrate a node that you skipped previously.
- If the stream you imported uses scripting, you may encounter an error when you run the flow even after completing a migration. This could be due to the flow script containing a reference to an unsupported import or export node. To avoid such errors, you must remove the scripting code that references the unsupported node.
- If the stream you're importing contains unsupported data file types, you need to convert them to a supported type (CSV, Excel, or SPSS Statistics .sav).
- In some cases, some settings from your original stream may not be restored during migration. For example, if the field delimiter in your original stream was tabs, it may be changed to commas after migration. Settings such as custom SQL also aren't migrated currently. Compare the new migrated flow to your original stream and making adjustments as needed.