ZFTurbo/Keras-inference-time-optimizer

KeyError on inbound_layers.append(res[inbound_layer_id])

rdaems opened this issue · 1 comments

I have a keyerror on init.py:49

This seems because model.layers does not include the input layer (I have a keras functional model).
I tried to solve it by changing model.layers to model._layers on line 34. But I have errors elsewhere now on a call to .get_layer().

Does kito support functional models?

Can you please create some small example based on your neural net which cause the error? I will fix it then.