If create the Linear() casually, it won't be trained during training.

[yangdechuan]

nlp-tutorial/5-1.Transformer/Transformer(Greedy_decoder)-Torch.py

Line 98 in 3b3a80d

output = nn.Linear(n_heads * d_v, d_model)(context)

[graykode]

Thanks for sharp caught Could you contribute for that? Also it contains jupyter file corresponding to code