Why would a ModelBuilder iteration throw Error 000918?

752
2
11-15-2019 09:54 AM
K_Rocks
New Contributor III

I had hoped to create a DTM raster using a series of point feature classes in a file geodatabase. Each feature class covers a grid with its points all lying 10m apart (all in a feature dataset with projected coordinate system). As there are almost 300 of them, I planned to use a ModelBuilder iteration to run through all the feature classes with the Point To Raster tool converting them. There is no problem inserting the Feature Classes iterator and the output shown in the ModelBuilder window appears to to pick up the feature classes fine. However, when I try to link this to a Point To Raster tool, I run into a problem.

Similar to this question and others I've found on StackExchange, I get an "Error 000918 - Cannot retrieve feature class" for the Input Features once I specify the desired output raster cellsize to be 10 in the Point To Raster tool set-up. This only arises using the iterator, though. I can run Point To Raster directly or in ModelBuilder for individual feature classes, setting the output cellsize as 10, without the error appearing. The input and output geodatabases are in different locations, so copying the features before Point To Raster, as suggested by some, doesn't make a difference.

Is there something obvious I'm missing here? Thanks for any help.

0 Kudos
2 Replies
curtvprice
MVP Esteemed Contributor

I think the issue is that the input points are inside a feature dataset, and the tool validation is having trouble with that and will not run. I suggest that after you get the path from the iterator, copy the points to an intermediate dataset (that is, NOT inside a feature dataset) with the Copy Features tool, then pass that copy to Points To Raster.

An alternative to copying that may work (but faster) is to pass the output of the iterator to the Make Feature Layer tool, which may wrap an extent that the Point To Raster tool will recognize. Worth a try.

Is this ArcMap or Pro?

0 Kudos
K_Rocks
New Contributor III

Thanks Curtis.

I never thought the feature dataset might be the problem but it appears that it is, just as you thought.

I've been using ArcMap 10.5.1 and tested things by copying a few of the feature classes to another geodatabase that doesn't have a feature dataset enclosing them. The model works fine then.