Hexagon Geospatial

Spatial Modeler

Discuss topics with other Hexagon Geospatial Product pioneers and experts to get the most out of our products.
Showing results for 
Search instead for 
Do you mean 
New Contributor
Posts: 3
Registered: ‎03-20-2019

Issue using Random Forest initialization



The SM Machine learning process is unable to solve the random forest initialization. I'm running Imagine 16.5.0 build 1265


The same model run with no issue on the same machine with the native 16.5 Imagine version


Was there any change on the expected machine configuration since the last Imagine version ?




01/04/19 18:38:00 SessionMgr(6032): Connection success for the external process 'eWkspace_64'
01/04/19 18:39:00 SessionMgr(6032): Running command line: C:/ERDAS2018/bin/x64URelease/mlpywrapper.exe "C:/ERDAS2018/etc/MachineLearning/Operators/randomforesttrain.py" "C:/Temp/SPATIAL_MODELER-cfe8-5da6-6d60-ab3c-005324/mlpy_jsoninput_7f076dbe-8059-4f7
a-9a6c-72f77c49e2be.Json" "C:/Temp/SPATIAL_MODELER-cfe8-5da6-6d60-ab3c-005324/mlpy_jsonoutput_ccfa71c4-65aa-42cb-9c6d-a17a6a33eb25.Json"
01/04/19 18:39:18 SessionMgr(6032): Could not understand JSON for type: IMAGINE.MachineIntellect
01/04/19 18:39:18 SessionMgr(6032): Operator execution failed.

Posts: 130
Registered: ‎06-30-2016

Re: Issue using Random Forest initialization

There is no change in system requirements among ERDAS IMAGINE 16.5, ERDAS IMAGINE 16.5 Update 1 and ERDAS IMAGINE 16.5 Update 2.


Regards, Sam

New Contributor
Posts: 3
Registered: ‎03-20-2019

Re: Issue using Random Forest initialization

Hi Sam



Thanks for your response. I've tested the script on two PC and obtained the same result. The initialization failed. However, I managed to run the same script with no issue in April 2018. I think something has been changed in the reading of the shape format specifications.







Do you need immediate support?
If you encounter a critical issue and need immediate assistance please submit a Service Request through our Support Portal.