I tried to modify the 09_Wide and deep learning on Census data Keras example, and received the error: “Invalid tensor shape. Expected shape dimension greater than 0, was 0 in dimension 0”. This occurred in the Keras Dense Layer node in the Model Creation metanode. Are there any ideas about what this error means, and how to correct it?
HI @Bob_Nisbet ,
The original example workflow runs fine, at least for me. The node that is falling for you is getting a network of shape  as an input from its predecessor (Keras Concatenate Layer) node. (See Screenshot)
Can you please check what you have in your case and share it here?
You can find out the dimension of the network coming out of a node, by right-clicking on the executed node and select Keras Network (last item in the menu).