Receiving ERROR 999999 while running Train Support Vector Machine

606
0
11-08-2021 07:04 PM
MattMoy
New Contributor II

I have been trying to run the Train Support Vector Machine to create a Classifier Definition File (.ecd) for a supervised image classification but keep receiving ERROR 999999. The input being used for the Train SVM tool is the output from previously running the Segment Mean Shift (SMS) tool. After running the SMS tool, I receive 2 outputs: a raster layer that uses the name I input into the SMS tool, and another raster layer with the identical name plus "_interIndex" added to the end of it.

I'm able to successfully run the Train SVM if the "fileName_interIndex" is used for the Input Raster in the Train Support Vector Machine tool and a Classifier Definition File is created. However, the results from the Classify Raster tool using this Classifier Definition File has large holes in the output raster.

If I use the file without the "_interIndex" in its name as the Input Raster for the Train Support Vector Machine tool, the tool fails and I receive the ERROR 999999. Does anyone have an explanation?

0 Kudos
0 Replies