Pipeline Case Iterator is a Slicer extension that allows for automated running of a pipeline over data in a csv file via a GUI interface.
- Select the pipeline you want to run in the
Pipeline Case Iteratormodule. - Select the csv file containing the cases you want to run the pipeline on. If this is the first time you are running the pipeline, you can use the Template Parameter File button to create a stub csv file, for types that are nodes an path to the file should be given.
- Select the output directory where you want the results to be saved. The output directory will contain all the results of the pipeline, as well as a
.csvfile with the appropriate input and output parameters for each row. - Optionally select an output extension for the files generated by the pipeline. A prefix and suffix and an timestamp can also be added to the output files.
- Press Run to run the pipeline on all the cases in the csv file.
The PipelineCaseIterator module allows for the automated running of a pipeline over a set of cases. The module requires a pipeline to be selected, a csv file containing the cases to be run, and an output directory where the results will be saved.
The Select button will bring up a a dialog to select the pipeline that you want to run. The inputs for the pipeline will be determined by a .cvs file containing the cases to be run.
Once you have set the output directory you are ready to run the pipeline. The output directory will contain all the results of the pipeline, as well as a .csv file with the appropriate input and output parameters for each row.
 |
|---|
| Case Iterator |
The input .csv file should contain a header row with the names of the parameters in the respective column. If you are unsure you can use the Create Template Parameter File button to create a stub .csv file with the appropriate headers. The file created via the button will not only name the parameter for the column but will also include the type of the respective parameter in the form <name>:<type>. The type is informational and not required when you label your own .csv file.
For any type except node types all you need to do is to fill in the values for the parameters in the respective columns. For node types you need to provide a path to the file containing the node. The path can be either absolute or relative to the location of the .csv file.
Pushing Run will begin executing the pipeline on each row of the .csv file. To do this a new instance of slicer will be created to provide a clean environment for running the pipeline. The progress of the pipeline will be displayed in the Overall Progress bar. Each full bar indicates one case being completed and the text in the bar will show <current case>/<total cases>. Calculation can be stopped by using the Cancel button.
Note that for each row, every column needs to be filled out for the case to run. The iterator will emit an error and skip a row if there are issues with parsing a value or loading a file the. The resulting output will only contain the cases that were run.
Run will be disabled until you've picked a pipeline, the input file is set and valid for the pipeline, and the output directory is set.
Any resulting node data will be stored in the output directory, all resulting scalars will be stored in the .csv file in the output directory (including paths to the appropriate files for each case)
The Output Prefix, Output Suffix, and Output Extension fields allow for the customization of the output files. The Output Prefix and Output Suffix will be added to the beginning and end of the output files respectively. With Add Timestamp checkbox enabled the current date and time will be added to the output files. In the Advanced section you can modify the format of the timestamp.