facebookresearch/UnsupervisedMT

Issues

• How can I modify the code to train may own dataset on specific language?

  #111 opened a year ago by Bayson-create
  0

• I have trouble when run get_data_enfr.sh

  #110 opened 2 years ago by china-lmfei
  0

• Setting the random seed does not result in same outputs across runs

  #109 opened 2 years ago by meganndare
  0

• transformer multihead attention scaling layer error

  #108 opened 3 years ago by skswldndi
  0

• How train PBSMT + NMT

  #82 opened 6 years ago by JxuHenry
  6

• How to run PBSMT +NMT ?

  #107 opened 3 years ago by electleaf
  0

• RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation: [torch.cuda.FloatTensor [14, 32, 1536]], which is output 0 of AddBackward0, is at version 2; expected version 1 instead. Hint: enable anomaly detection to find the operation that failed to compute its gradient, with torch.autograd.set_detect_anomaly(True).

  #106 opened 4 years ago by rashikumar01
  1

• About number of shared layers

  #105 opened 4 years ago by pangjh3
  0

• UnboundLocalError: local variable 'n_words' referenced before assignment

  #104 opened 4 years ago by electleaf
  0

• How to train NMT + PBSMT ?

  #103 opened 4 years ago by electleaf
  0

• Why codes file is empty.?

  #101 opened 4 years ago by ykkhan
  4

• for different language, where to make change?

  #102 opened 4 years ago by ykkhan
  0

• why MemoryError

  #100 opened 4 years ago by ykkhan
  0

• Error in runny bash command.

  #99 opened 4 years ago by ykkhan
  2

• How to train the model without para_dataset

  #97 opened 4 years ago by huyphan168
  0

• How i can run MUSE alignment in .sh

  #96 opened 4 years ago by huyphan168
  0

• fastBPE

  #89 opened 5 years ago by ErolOZKAN-
  1

• Getting raise EOFError() while executing Linux Command through Netmiko

  #95 opened 4 years ago by techiaman1004
  0

• Bus error (core dumped)

  #90 opened 5 years ago
  1

• How can I solve the AssentionError?

  #69 opened 6 years ago by Lyabo
  4

• Mac can't go through line 149 in "get_data_*.sh".

  #72 opened 5 years ago by yongchanghao
  1

• Reproducing PBSMT german-english

  #64 opened 6 years ago by mayhewsw
  16

• bpe_end issue

  #94 opened 5 years ago by 17614040741
  0

• Why I get low BLEU on zh-en of NMT ?

  #83 opened 6 years ago by JxuHenry
  7

• Broken pipe from multiprocessing

  #87 opened 5 years ago by thammegowda
  3

• fastBPE: bad_alloc() Core dumped

  #59 opened 6 years ago by rucha3
  3

• Low BLEU on PBSMT

  #93 opened 5 years ago by kellymarchisio
  3

• How to train the vector of phrases

  #92 opened 5 years ago by liujiqiang999
  0

• Low utilization rate of cuda

  #91 opened 5 years ago by xzdong-2019
  1

• Ran out of tokens

  #88 opened 5 years ago by HoongLik
  0

• Train NMT with my own dataset

  #86 opened 5 years ago by jackson1895
  1

• Error duing pre-processing stage: en_es

  #67 opened 6 years ago by sedflix
  1

• Train zh-en:assert sorted(params.langs) == params.langs

  #78 opened 5 years ago by JxuHenry
  8

• How long does NMT need to run

  #73 opened 6 years ago by JxuHenry
  4

• interactive mode and/or run-time querying

  #65 opened 6 years ago by bittlingmayer
  1

• How to train chinese-english

  #74 opened 6 years ago by JxuHenry
  0

• how to train my data without pre-trained word embeddings?

  #70 opened 6 years ago by prigioni
  1

• How long would it take for you to train fastText?

  #81 opened 6 years ago by yongchanghao
  2

• Changes in FastBPE

  #80 opened 6 years ago by AlAntonov
  1

• How to slove this problem load_mono_data assert data['dico'][lang] == mono_data['dico'] AssertionError

  #75 opened 6 years ago by JxuHenry
  6

• Why use "assert sorted(params.langs) == params.langs" in the loader.py file at line 314

  #79 opened 6 years ago by JxuHenry
  1

• AssertionError running the run.sh file

  #77 opened 6 years ago by erthher
  1

• Training NMT: How to solve _pickle.UnpicklingError: invalid load key, 'D'.

  #76 opened 6 years ago by erthher
  0

• Using this code (transformer) on Multi30k English French monolingual

  #57 opened 6 years ago by ahmadrash
  5

• Add a Dockerfile for this project

  #61 opened 6 years ago by sedflix
  1

• I have got an error when run.sh

  #71 opened 6 years ago by prigioni
  6

• Question about EN-RU

  #66 opened 6 years ago by cocaer
  1

• How to deal with casing?

  #62 opened 6 years ago by mayhewsw
  2

• Does the paper result uses beam search?

  #63 opened 6 years ago by cocaer
  2

• CUDA out of memory during test set evaluation only

  #58 opened 6 years ago by dkavaler
  1