DLS 5 week 4 exercise 4

Course Q&A Deep Learning Specialization Sequence Models
1

/opt/conda/lib/python3.7/site-packages/tensorflow/python/keras/engine/base_layer.py in _split_out_first_arg(self, args, kwargs)
3046 else:
3047 raise ValueError(
→ 3048 ‘The first argument to Layer.call must always be passed.’)
3049 return inputs, args, kwargs
3050

ValueError: The first argument to Layer.call must always be passed.

I cannot figure out what the syntax should be

2

samee even i am stuck here, it’s soo frustrating ugh

3

If this error happens at the full encoder (UNQ_C5), then it is because the the full encoder “calls” the EncoderLayer. If EncoderLayer test is successfully passed, the MultiHeadAttention layer should have been implemented properly in the call function there. So the job left for the call function in the full encoder is to call the EncoderLayer with the proper arguments.