Confused by UNQ_C1 instruction:
Use repeator to repeat s_prev to be of shape (m, Tx, n_s) so that you can concatenate it with all hidden states “a” (≈ 1 line)
Where Keras documentation notes syntax:
RepeatVector( n, **kwargs)
I’ve tried various arguments with repeator( ) and have gotten the following error messages:
TypeError: ‘tensorflow.python.framework.ops.EagerTensor’ object is not callable
ValueError: The first argument to Layer.call must always be passed. {got a number of these error messages for a variety of reasons}
AttributeError: ‘tuple’ object has no attribute ‘shape’
AssertionError:
Given repeator is define as:
repeator = RepeatVector(Tx)
it would seem to be simply passing repeator s_prev BUT no.
Clear guidance requested…thanks!
Thanks for the pointer, but I’ve tried 7 or 8 syntactical variations with both functional and sequential syntax…still error messages.
Given I’ve tried so many variations I now don’t know why any of them won’t work.
Not sure how to help you help me…it’s been at least 2 hours of working on this one statement…
I do have all the assignment diagrams and instructions in separate images right next to the notebook on a 48-inch monitor–in 5 courses this is the first time I haven’t been able to start an exercise…
Solved the problem by renaming the assignment and getting the latest version of the file.
I guess any time I’m having problems with something with “global” vars, my first actions should be getting an updated version and starting over (not a problem here because I was stuck as step 1).