Regarding the undefined attention_weights in the train_symbol_decoder.ipynb file

[anonymouscoder1022]

Hi,

Thank you for your work, I’ve learned a lot from it. However, in the train_symbol_decoder.ipynb file, attention_weights in cell 58 is not defined, which causes the code to fail. Could you please check this?

Additionally, sentence in cell 59 and translated_tokens in cells 60 and 61 are also not defined.

head = 0
# Shape: `(batch=1, num_heads, seq_len_q, seq_len_k)`.
attention_heads = tf.squeeze(attention_weights, 0)
attention = attention_heads[head]
attention.shape

Thanks in advance!

[tunaozates]

Thank you for your interest in our work and also helping us to fix the issue! We now updated the train_symbol_decoder.ipynb so that it runs without errors.