harvardnlp/seq2seq-attn

Issues

• Segmentation fault

  #109 opened 2 years ago by reBiocoder
  0

• code for seq-level distillation

  #100 opened 7 years ago by Oneplus
  1

• when i try to train model gets below error----hdf5

  #90 opened 7 years ago by akshataph
  4

• Issue when using Guided Alignment

  #106 opened 6 years ago by Anupama94
  0

• multi-GPU

  #105 opened 7 years ago by wanghechong
  0

• What exactly does `-use_chars_dec` do?

  #104 opened 7 years ago by arendu-zz
  0

• Error when training

  #103 opened 7 years ago by anbasile
  0

• extract_states.lua's error

  #97 opened 7 years ago by zhang-jinyi
  0

• Sequence-to-Sequence Learning as Beam-Search Optimization

  #99 opened 7 years ago by jndevanshu
  2

• I got an idea....

  #98 opened 7 years ago by SeekPoint
  0

• How to get the context vector

  #94 opened 7 years ago by lenhhoxung86
  2

• how to use s2sa/scorers/bleu.lua

  #93 opened 7 years ago by IdiosyncraticDragon
  1

• How to compute the probability for a pair of input and output sentences?

  #96 opened 7 years ago by ayushidalmia
  2

• model's PPL

  #95 opened 7 years ago by zhang-jinyi
  4

• Preprocess failed for the WMT'14

  #92 opened 8 years ago by IdiosyncraticDragon
  4

• Can prune.lua apply to the model trained by OpenNMT

  #91 opened 8 years ago by IdiosyncraticDragon
  2

• Implementing biased decoding and tagging words

  #88 opened 8 years ago
  5

• How did you deal with exposure bias in your implementation?

  #89 opened 8 years ago by blackyang
  2

• Character vectors

  #50 opened 8 years ago by alashkari
  3

• preprocess-shards compatibility with train

  #77 opened 8 years ago by hsajjad
  1

• Have you try to add linguistic features in your nmt experiment? I need some help.

  #85 opened 8 years ago by hfxunlp
  6

• Character CNN model on decoder side

  #86 opened 8 years ago by boknilev
  2

• dataset credits

  #87 opened 8 years ago by christopher5106
  1

• input feed

  #84 opened 8 years ago by christopher5106
  1

• Error in fine tuning with 2 gpus

  #83 opened 8 years ago by ankitag9
  2

• Ask for help

  #80 opened 8 years ago by hfxunlp
  4

• What is the input of decoder?

  #79 opened 8 years ago by hfxunlp
  2

• why I cannot use gpu 0?

  #64 opened 8 years ago by SeekPoint
  3

• 0d source

  #62 opened 8 years ago by mzhai2
  1

• train_from not working for old models

  #56 opened 8 years ago by ankitag9
  7

• Memory requirements in preprocessing

  #72 opened 8 years ago by oraveczcsaba
  2

• total_loss_cll with nil value

  #78 opened 8 years ago by oraveczcsaba
  1

• Trouble with Spatial Convolution

  #76 opened 8 years ago by alashkari
  4

• Tiny bug in preprocessing.py

  #75 opened 8 years ago by helson73
  3

• About source batch lengths

  #74 opened 8 years ago by helson73
  2

• maybe a bug

  #73 opened 8 years ago by codyhan
  3

• Determine pre-allocated storage

  #71 opened 8 years ago by helson73
  3

• Memory pre-allocation in LSTM

  #70 opened 8 years ago by helson73
  4

• Error on training when used preprocess_shards

  #69 opened 8 years ago
  2

• Issue with evaluate script.

  #68 opened 8 years ago by randomizedai
  3

• Recurrent highway network

  #67 opened 8 years ago by christopher5106
  0

• Train from logical error

  #66 opened 8 years ago by hanskrupakar
  6

• out of memory during the training

  #65 opened 8 years ago by jdxyw
  3

• bad argument #2 to '?' (end index out of bound) error

  #63 opened 8 years ago by hanskrupakar
  5

• Visualize attention weights

  #58 opened 8 years ago by fbrad
  3

• error loading module 's2sa.models' ... ambiguous syntax (function call x new statement)

  #54 opened 8 years ago by ili3p
  2

• Using this for sorting

  #47 opened 8 years ago by ch3njust1n
  1

• Data parallel training

  #59 opened 8 years ago by hrishikeshvganu
  1

• Create Character level data-set using shards

  #43 opened 8 years ago by mdasadul
  7

• Understanding the data preprocessing

  #44 opened 8 years ago by vikram-gupta
  6