Hi, I'm running a random forest classification using the Spatial Statistics Tool "Forest Based and Boosted Classification and Regression." I have three explanatory 30m rasters (canopy height, canopy layers, large tree density), all of which are numeric.
Prediction Type: Predict to raster
Model Type: Forest-based
Input training features: a point feature class with a field that indicates presence and absence (True/False, so I am treating the variable as categorical) of mature forest across the area of interest.
The tool seems to nearly finish running, but fails with the following errors. Has anyone seen the same errors before or know what they indicate?
Traceback (most recent call last):
File "c:\program files\arcgis\pro\Resources\ArcToolbox\scripts\SSForest.py", line 8124, in <module>
#### Projection Group ####
^^^^^^^^^^^^^^^^^^^
File "c:\program files\arcgis\pro\Resources\ArcToolbox\scripts\SSForest.py", line 7556, in execute
self.yMax = self.intersection[3]
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "c:\program files\arcgis\pro\Resources\ArcToolbox\scripts\SSForest.py", line 7455, in rasterWorkFlow
def getWorkExtent(self):
^^^^^^^^^^^^^^^^
File "c:\program files\arcgis\pro\Resources\ArcToolbox\scripts\SSForest.py", line 5658, in executeModel
return None
^^^^^^^^^^^^
File "c:\program files\arcgis\pro\Resources\ArcToolbox\scripts\SSForest.py", line 5520, in processRasterByParts
outputStats = outputStats.reshape((len(indices), numIterationRobust))
^^^^^^^^^^^^^^^^^^
AttributeError: 'NoneType' object has no attribute 'shape'
Failed script Forest-based and Boosted Classification and Regression...
Failed to execute (Forest).
