harvardnlp seq2seq-attn issues

harvardnlp / seq2seq-attn

Sequence-to-sequence model with LSTM encoder/decoders and attention

http://nlp.seas.harvard.edu/code

MIT License

1.26k stars 278 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Segmentation fault

#109 reBiocoder opened 2 years ago
0
Issue when using Guided Alignment

#106 Anupama94 opened 6 years ago
0
multi-GPU

#105 wanghechong opened 6 years ago
0
What exactly does `-use_chars_dec` do?

#104 arendu-zz opened 6 years ago
0
Error when training

#103 anbasile opened 6 years ago
0
Add nil check to avoid 'table index is nil' errors

#102 twistedcubic opened 6 years ago
0
Fix require statement to avoid module not found errors

#101 twistedcubic opened 6 years ago
0
code for seq-level distillation

#100 Oneplus opened 6 years ago
1
Sequence-to-Sequence Learning as Beam-Search Optimization

#99 jndevanshu closed 7 years ago
2
I got an idea....

#98 SeekPoint opened 7 years ago
0
extract_states.lua's error

#97 zhang-jinyi closed 6 years ago
0
How to compute the probability for a pair of input and output sentences?

#96 ayushidalmia closed 7 years ago
2
model's PPL

#95 zhang-jinyi closed 7 years ago
4
How to get the context vector

#94 lenhhoxung86 closed 7 years ago
2
how to use s2sa/scorers/bleu.lua

#93 IdiosyncraticDragon closed 7 years ago
1
Preprocess failed for the WMT'14

#92 IdiosyncraticDragon closed 7 years ago
4
Can prune.lua apply to the model trained by OpenNMT

#91 IdiosyncraticDragon closed 7 years ago
2
when i try to train model gets below error----hdf5

#90 akshataph closed 7 years ago
4
How did you deal with exposure bias in your implementation?

#89 blackyang closed 7 years ago
2
Implementing biased decoding and tagging words

#88 ghost closed 7 years ago
5
dataset credits

#87 christopher5106 closed 7 years ago
1
Character CNN model on decoder side

#86 boknilev closed 7 years ago
2
Have you try to add linguistic features in your nmt experiment? I need some help.

#85 hfxunlp closed 7 years ago
6
input feed

#84 christopher5106 closed 7 years ago
1
Error in fine tuning with 2 gpus

#83 ankitag9 closed 7 years ago
2
Command-line option for layer-specific learning rates for non-sgd optimization

#82 swiseman closed 7 years ago
0
Disable CUDA cross-GPU checks when 2 GPUs are used

#81 macournoyer opened 7 years ago
0
Ask for help

#80 hfxunlp closed 7 years ago
4
What is the input of decoder?

#79 hfxunlp closed 7 years ago
2
total_loss_cll with nil value

#78 oraveczcsaba closed 7 years ago
1
preprocess-shards compatibility with train

#77 hsajjad closed 7 years ago
1
Trouble with Spatial Convolution

#76 alashkari closed 7 years ago
4
Tiny bug in preprocessing.py

#75 helson73 closed 7 years ago
3
About source batch lengths

#74 helson73 closed 7 years ago
2
maybe a bug

#73 codyhan closed 7 years ago
3
Memory requirements in preprocessing

#72 oraveczcsaba closed 7 years ago
2
Determine pre-allocated storage

#71 helson73 opened 7 years ago
3
Memory pre-allocation in LSTM

#70 helson73 closed 7 years ago
4
Error on training when used preprocess_shards

#69 ghost closed 8 years ago
2
Issue with evaluate script.

#68 randomizedai closed 8 years ago
3
Recurrent highway network

#67 christopher5106 opened 8 years ago
0
Train from logical error

#66 hanskrupakar opened 8 years ago
6
out of memory during the training

#65 jdxyw closed 8 years ago
3
why I cannot use gpu 0?

#64 SeekPoint closed 7 years ago
3
bad argument #2 to '?' (end index out of bound) error

#63 hanskrupakar closed 8 years ago
5
0d source

#62 mzhai2 closed 7 years ago
1
Attention retrieval added

#61 sebastianGehrmann closed 8 years ago
1
Fallback to default options when training from old models

#60 guillaumekln closed 8 years ago
0
Data parallel training

#59 hrishikeshvganu closed 8 years ago
1
Visualize attention weights

#58 fbrad closed 8 years ago
3