nyu-dl dl4mt-tutorial issues

nyu-dl / dl4mt-tutorial

BSD 3-Clause "New" or "Revised" License

618 stars 249 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

TypeError

#81 sourabhdeoghare opened 5 years ago
0
Can anyone provide GPU version of translate.py?

#80 jonghkim opened 6 years ago
1
By condition what does it mean in this tutorial?

#79 eduOS closed 7 years ago
0
Maybe it's time to upgrade to Python 3 and ditch Python 2 support?

#78 jli05 opened 7 years ago
1
compute word probabilities

#77 hanyaqian closed 7 years ago
2
the value of TensorType(float32, 3D)

#76 hanyaqian closed 7 years ago
2
a detailed description about param_init_gru and gru_layer

#75 hanyaqian closed 7 years ago
2
why convert the value of matrix to the type with astype('float32')?

#74 hanyaqian closed 7 years ago
2
ValueError: unsupported pickle protocol: 3

#73 hanyaqian closed 7 years ago
2
Update data preprocessing for Python 3

#72 jli05 closed 7 years ago
4
Random Translations?

#71 Lotemp closed 8 years ago
5
How to build the dataset 'all.en.concat.gz.pkl' in session2/train_nmt_all.py?

#70 wead-hsu closed 8 years ago
1
dim == dim_nonlin and nin == nin_nonlin must be always true?

#69 jozef-mokry opened 8 years ago
0
Asymmetry in read gate application

#68 jozef-mokry opened 8 years ago
0
Unnecessary bias term?

#67 jozef-mokry closed 8 years ago
1
L2 regularization on bias terms?

#66 jozef-mokry opened 8 years ago
0
fix bug in gru layer

#65 michaelpulsewidth closed 8 years ago
6
NaN detected

#64 amirj opened 8 years ago
1
discrepancy between paper and code

#63 ethancaballero closed 8 years ago
1
Where is dataset='/ichec/work/dl4mt_data/nec_files/wiki.tok.txt.gz'?

#62 JunjieHu opened 8 years ago
2
Added import of theano.shared in translate.py

#61 ofirpress closed 8 years ago
1
Cost is Nan after one epoch if maxlen > 50

#60 ethancaballero closed 8 years ago
3
Avmb branch

#59 Avmb closed 8 years ago
0
Why the grads have to be shared?

#58 luochuwei closed 8 years ago
2
Latest adam with recommended parameters

#57 bastings closed 8 years ago
1
Do you have any plan for the next deep learning for machine translation school?

#56 amirj closed 8 years ago
2
what is the role of 'maxlen' parameter?

#55 amirj opened 8 years ago
3
use standard adam parameters

#54 bastings closed 8 years ago
15
documentation

#53 orhanf closed 8 years ago
0
What's the role of 'dim_word' parameter in the 'train()' method?

#52 amirj closed 8 years ago
2
README missing

#51 amartya18x closed 8 years ago
0
Would it be possible to release code in Python 3.4+ for any future repos?

#50 jli05 closed 8 years ago
7
I can't re-run a previous perfect train_nmt.py

#49 amirj closed 8 years ago
2
Output contains lot of UNKs

#48 arvind2505 closed 8 years ago
11
Preparing dataset for neural MT

#47 arvind2505 closed 8 years ago
2
How I can run the language model

#46 jifan-chen closed 8 years ago
2
Fixes decoder projection layer indexing bug in Session 1

#45 scfrank closed 8 years ago
0
Small problem in merge.sh

#44 aalmah closed 8 years ago
1
On decoder to softmax layer projection in session 1

#43 ustctf-zz closed 8 years ago
2
Refactoring

#42 jli05 closed 8 years ago
0
nmt: typo in variable names

#41 ozancaglayan closed 8 years ago
1
Multi-Task Learning for Multiple Language Translation

#40 amirj closed 8 years ago
2
fix use_noise

#39 kelvinxu closed 8 years ago
0
"reload = true" behavior ?

#38 vince62s closed 8 years ago
4
MT corpus and approach

#37 jli05 closed 8 years ago
4
UNK replacement

#36 vince62s closed 8 years ago
3
saving model with iteration number for session1,2,3 + sync session2-3

#35 orhanf closed 8 years ago
0
hardcoded paths need to be fixed in nmt.py or train_nmt.py or both ?

#34 vince62s closed 8 years ago
1
How to define a task-specific validation measure?

#33 amirj closed 8 years ago
7
Session 1

#32 vince62s closed 8 years ago
9