Viewing job details for a data mining run

The Job Detail page provides a list of parameters used for a component job in a data mining run, and a list of input and output files produced by the selected job. Certain files are always saved and available from this page, while other files are saved and available only if the run creator checked Save intermediate data files on the Run Options page.

• An error log file called job_job-id_log-id.log is produced for each extract or LR_extract job. This log contains the date and time that error logging began and messages about any errors that occurred. For LR runs, this log also includes a line for each process in the run.
• The lr_coefficients.log file is produced for each LR job. This file contains the computed best alpha value for each response in an extended LR computation, and also includes the coefficient and standard error values computed for each predictor-response combination if the run creator checked Save coefficients on the Run Options page.
• The lr.log file is produced for each LR job. This file identifies the most common value for each covariate in the run and summarizes convergence for each response.
• For the mgps job, the datamine.log file shows the version of the MGPS engine that you used. For the LR job in a logistic regression run, the lr.log file shows the version. The version number is helpful to know when communicating with Oracle about issues related to the data mining run.

1.         On the Data Mining Runs page, click the row menu () for a run, and then click View Jobs for Run.

The Jobs for Run page appears.

2.         From the Name column, click the job.

The Job Detail page appears.

3.         To download a file, from the Job Parameters section, click the input or output filename.

The Windows File Download dialog box appears.

4.         You can open the file or save it. If the file is zipped, you must have a ZIP file compression and extraction utility, such as WinZip, installed on your computer, as described in Prerequisites and usage Notes.