C5w4 A1 exercise 6 DecoderLayer()

Hello guys,

Can’t figure out what I did wrong when calling self.mha2. Could you please assist?

I call it with the set of arguments (Q1, enc_output, enc_output, attention_mask = look_ahead_mask, return_attention_scores=True) exactly as it is.

1 Like

mha2 will use the ‘padding_mask’ and not the ‘look ahead mask’. I think this solves it for it.

2 Likes

Yes, @vsnupoudel, you are right.

It solved the problem, thank you.