neulab/awesome-align

Issues

• Setting seq-to-seq model as our pretrained model

  #49 opened 2 years ago by b3ade
  6

• Repeated single-sentence inferences on an in-memory model?

  #58 opened a year ago by pseudomonas
  1

• Training details

  #57 opened 2 years ago by CBHhD
  2

• Torch.save() for large training Dataset

  #56 opened 2 years ago by stelmath
  1

• A bug(maybe)

  #55 opened 2 years ago by shaoyangxu
  1

• Inputs shoult be tokenized only for training/evaluation sets?

  #54 opened 2 years ago by stelmath
  1

• Is it possible align phrase by phrase, not just word by word?

  #52 opened 2 years ago by fishfree
  4

• Is it possible to incorporate POS tag info to aid alignment?

  #53 opened 2 years ago by stelmath
  1

• Question on Figure 2 in your paper

  #51 opened 2 years ago by gpengzhi
  2

• Extracting dataset and AttributeError

  #44 opened 2 years ago by b3ade
  3

• Alignment prob matrix visualization tool?

  #50 opened 2 years ago by kirklandWater1
  2

• Trying to train "ixa-ehu/ixambert-base-cased" model

  #48 opened 2 years ago by jmurua14
  1

• Trying to train on an existing model

  #47 opened 2 years ago
  1

• Statistics between models

  #46 opened 2 years ago by b3ade
  1

• traning is increasing my aer

  #36 opened 2 years ago by vigneshmj1997
  3

• DistributedDataParallel does not work on some PyTorch versions

  #13 opened 2 years ago by zdou0830
  2

• How to test fine tuned model on parallel data?

  #40 opened 2 years ago by MusfiqDehan
  3

• How can I resume training from a checkpoint?

  #42 opened 2 years ago by MusfiqDehan
  1

• Abort experiment in Table 8

  #45 opened 2 years ago by sdongchuanqi
  3

• How to get test results?

  #43 opened 3 years ago by QzzIsCoding
  2

• Continue with checkpoint

  #39 opened 3 years ago by kirklandWater1
  2

• Parallel corpus data format for fine-tuning on parallel data

  #38 opened 3 years ago by shihabshahriar16
  1

• 一些问题

  #37 opened 3 years ago by lisasiyu
  2

• Can I evaluate separately?

  #34 opened 3 years ago by ruoyuxie
  1

• Wrong mapping with non-matching sentences

  #9 opened 4 years ago by mzeidhassan
  5

• problem with mask_token

  #32 opened 3 years ago by v-zixin
  3

• Training with gold alignment

  #31 opened 3 years ago by ruoyuxie
  2

• There is a small problem about the training data format.

  #29 opened 3 years ago by genbei
  1

• Fasten the extraction process

  #30 opened 3 years ago by shanyas10
  2

• AttributeError: Can't pickle local object 'word_align.<locals>.collate'

  #28 opened 3 years ago by manandey
  4

• De-En dataset is missing

  #27 opened 3 years ago by ruoyuxie
  1

• setting output prob file and output word file

  #26 opened 3 years ago by pskadasi
  2

• Can I use the XML-R model?

  #23 opened 3 years ago by sadanyh
  5

• Can I use awesome-align for monosyllabic language

  #25 opened 3 years ago by quocthang0507
  2

• 一些问题

  #22 opened 3 years ago by leileilin
  4

• Not able to reproduce the AERs in the table

  #20 opened 3 years ago by jinyiyang-jhu
  2

• Relationship between num_train_epochs and max_steps

  #18 opened 3 years ago by dmitrytoda
  2

• some confusions

  #17 opened 3 years ago by leileilin
  8

• package `requests` missing from the requirements file

  #16 opened 4 years ago by antonisa
  2

• Reproduction of Ja-En

  #14 opened 4 years ago by twadada
  6

• No caching

  #11 opened 4 years ago by BramVanroy
  5

• Multiple GPUs

  #10 opened 4 years ago by BramVanroy
  0

• pip packaging?

  #8 opened 4 years ago by neubig
  4

• Packaging

  #5 opened 4 years ago by bitPogo
  4

• Index out of range

  #3 opened 4 years ago by mapmeld
  10

• Error reading fine-tuned models

  #2 opened 4 years ago by dmitrytoda
  2

• UnicodeDecodeError: 'utf-8' codec can't decode byte 0x80 in position 0: invalid start byte

  #1 opened 4 years ago by echan00
  1