I have successfully trained a feature classification model with arcgis deep learning toolbox using jupyter notebook and want to do inference with my new model on a new zone. I am using google map extracted high resolution satellite imafes. But using the ClassifyObjectsUsingDeepLearning tool always ends up in
ERROR 999999: Something unexpected caused the tool to fail. Contact Esri Technical Support (http://esriurl.com/support) to Report a Bug, and refer to the error help for potential solutions or workarounds. Failed to execute (ClassifyObjectsUsingDeepLearning).
It happens both when I use the tool inside the GIS (Pro 2.9.1) and also through jupyter and arcpy. It also happens regardless of using GPU or CPU. I really have no clue what happens... I cannot contact esri support because I am on a student license.
Thank you very much for your consideration
Classify Objects Using Deep Learning (Image Analyst)—ArcGIS Pro | Documentation
Classify Objects Using Deep Learning (Raster Analysis)—ArcGIS Pro | Documentation
both access points have specific requirements and assume that the proper environment has been set up
Esri/deep-learning-frameworks: Installation support for Deep Learning Frameworks for the ArcGIS Syst...
are you running the process from a cloned environment? or is everything in the base environment?
Dear Dan, thank you so much for your answer
I think proper environment is well set up from the documentation indications.
The latest version of the deep learning framework is also used and worked well to train my model.
I am using everything in the base environment after I installed deep learning framework
It is the "...it contains the path to... " stuff in the parameters that often calls things to fail, I would just check your details, because the 999999 message is almost a declaration of "... no clue what is wrong ..."
You can try searching similar questions in this community or
Problem: Unable to execute the Detect Objects Using Deep Learning tool in ArcGIS Pro (esri.com)
you can spend the night reading similar messages on the Support site.