keon seq2seq issues - Githubissues

keon / seq2seq

Minimal Seq2Seq model with Attention for Neural Machine Translation in PyTorch

MIT License

689 stars 172 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

A question about the nn.Embedding

#30 zhang-qiang-github opened 3 years ago
0
Fix incorrect argument description

#29 CirQ opened 3 years ago
0
Why using relu to compute additaive attention

#28 yuboona opened 4 years ago
0
fix(evaluate): use no_grad over volatile=True

#27 AmitMY closed 4 years ago
0
fix(train): use non-deprecated grad clip

#26 AmitMY closed 4 years ago
0
fix(loss): get value of 0-dim tensor using `tensor.item()`

#25 AmitMY closed 4 years ago
0
fix attention

#24 keon closed 4 years ago
0
A problem with loss computation.

#23 yxdr opened 4 years ago
1
don't have the inference mode?

#22 ArnoldLIULJ opened 5 years ago
0
modify relu function

#21 kenya-sk closed 5 years ago
0
[EROOR] Not Work Relu

#20 kenya-sk closed 5 years ago
2
Repo is broken

#19 CuriousG102 closed 4 years ago
1
Hello,I have a question about the "encoder_outputs"of the model.py

#18 Biubiulity opened 5 years ago
0
Hello,I have a question about the

#17 Biubiulity opened 5 years ago
0
in model.py line 76: context = attn_weights.bmm(encoder_outputs.transpose(0, 1)) # (B,1,N)

#16 Huijun-Cui closed 3 years ago
1
about the way to calculate attention weight

#15 FreyWang opened 5 years ago
2
About overfitting

#14 Linao1996 opened 5 years ago
1
Update train.py

#13 Linao1996 closed 5 years ago
0
What's the exact Pytorch and Torchtext version for your code? I am trying to downgrade to a previous version in order to avoid the Multi30k.split() problem but failed.

#12 yaoyiran opened 5 years ago
5
torchtext Multi30k

#11 NiceMartin opened 6 years ago
2
The Pytorch version？

#10 ghost opened 6 years ago
1
No teacher forcing during evaluation

#9 pskrunner14 closed 6 years ago
3
Model still uses teacher forcing when evaluating

#8 ruizheliUOA closed 6 years ago
1
what is the nn.Parameter v for?

#7 liuyijiang1994 closed 6 years ago
3
Non linearity after attention

#6 pskrunner14 closed 6 years ago
3
Crossentropy to NLL Loss

#5 pskrunner14 closed 6 years ago
1
A bug of the Loss function in the def 'train' and 'evaluate'

#4 ruizheliUOA closed 6 years ago
3
Use of non-linearity in calculating attention

#3 tarun005 closed 6 years ago
1
random seed

#2 dalinvip closed 6 years ago
0
About the usage of initial hidden state in calculating attention

#1 leitro closed 6 years ago
2