Error with next_symbol (UNQ_C6)

Course Q&A Natural Language Processing NLP with Attention Models
1

Hi,

I’m getting the following error when unit testing UNQ_C6.

May labid: myaearqmtwuu

LayerError: Exception passing through layer Serial (in pure_fn):
layer created in file […]/, line 64
layer input shapes: ShapeDtype{shape:(1, 5), dtype:int64}

File […]/trax/layers/base.py, line 707, in setattr
super().setattr(attr, value)

File […]/trax/layers/base.py, line 454, in weights
f’Number of weight elements ({len(weights)}) does not equal the ’

ValueError: Number of weight elements (1) does not equal the number of sublayers (9) in: Serial_in2_out2[
Select[0,1,0,1]_in2_out4 …

I’ve checked the input’s dimensions and the padding and I don’t understand why I’m getting this error. A side question: I’m only padding the cur_output_tokens as suggested, but shouldn’t I be padding the input_tokens as well ?
Could you please help me to understand why do I have this problem ?
Thanks !
Francisco

2

Hi @Francisco_Romero I am not sure where the problem is but I think you should chack and redo again the "# UNQ_C4

GRADED FUNCTION

def NMTAttn"

function, I think the problem might be there!

3

Hi @gent.spah ,

Thanks for the feedback. I’ve gone over #UNQ_C4, but I don’t see what I’m missing. All tests are passing and the printed model expected result is identical to what I get.

4

Hi @gent.spah ,

Found the problem. I had an error syntax when calling the model in UNQ_C6. It works now. Thanks !
Francisco