I think I misread the tutorial, I'm supposed to pass in the uncompiled function right?
On Monday, September 26, 2016 at 7:34:41 PM UTC+1, Andy Jiang wrote: > > following this -> > http://deeplearning.net/software/theano/tutorial/printing_drawing.html > > I have theano.printing.pprint(prediction) before training to show the > graph, where > > prediction = theano.function( > inputs=[x], > outputs=classifier.output, > name='predict' > ) > > here classifier is a neural network. I am getting the error > > File "---theano\theano\printing.py", line 582, in __call__ > return self.process(*args) > File "---theano\theano\printing.py", line 517, in process > if condition(pstate, r): > File "---theano\theano\printing.py", line 507, in <lambda> > condition = (lambda pstate, r: r.owner is not None and > AttributeError: 'Function' object has no attribute 'owner' > > If needed I can provide more details on implementation of classifier. > > How do I fix this? > > -- --- You received this message because you are subscribed to the Google Groups "theano-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
