C4W2 - InvalidArgumentError: Exception encountered when calling layer 'softmax_157'

[Deepti_Prasad]

hi @Ernest_Divine

check the below thread if it helps you debug your codes.

One of the issue clearly seen in your code is you have used training for block 2, which you are not suppose to.

Also check your output code and the padding mask recall to the output. if they are correctly recalled

NLP with Attention Models C4W2 - Exercise 5 - next_word “softmax_404” error

Hi @John_Murphy1 Issues with your assignment as per grade cell GRADED FUNCTION: scaled_dot_product_attention 1.softmax is normalized on the last axis (seq_len_k) so that the scores add up to 1. YOU DONT REQUIRE AXIS FOR THIS CODE RECALL GRADED FUNCTION: DecoderLayer For Block 1, it is clearly mentioned Dropout will be applied during training, so you do not require training=training for your self-attention mult_attn_out1. Same issue with mult_attn_out2 GRADED FUNCTION: Decoder No mistake…

in case you still not able to find, let me know.

Regards
DP