Unbabel/OpenKiwi

Issues

• What languages do the OpenKiwi support?

  #116 opened 7 months ago by wangpichao
  0

• Why does it need "--model" paramter when I give a specific config?

  #115 opened 2 years ago by michelleqyhqyh
  2

• What are source_pos and target_pos in the train_config.yaml?

  #114 opened 2 years ago by zolastro
  0

• Pretrain config file

  #113 opened 2 years ago by zolastro
  0

• Do you need to tokenize your data when using a BERT/ROBERTA model?

  #111 opened 3 years ago by zolastro
  0

• PicklingError: Can't pickle <class 'kiwi.data.encoders.wmt_qe_data_encoder.InputFields[PositiveInt]'>: attribute lookup InputFields[PositiveInt] on kiwi.data.encoders.wmt_qe_data_encoder failed

  #110 opened 3 years ago by Blinkblade
  2

• Seems that maximum token support for a sentence is 512?

  #109 opened 3 years ago by Duonghailee
  0

• open cannot unpack non-iterable NoneType object

  #99 opened 3 years ago by genbei
  16

• Got exception when import kiwi

  #108 opened 3 years ago by gaohuan2015
  0

• pkgutil.iter_modules() error: 'PosixPath' object has no attribute 'startswith'

  #107 opened 3 years ago by mtreviso
  0

• some confusions

  #106 opened 3 years ago by leileilin
  0

• OpenKiwi always download the tokenizer files for XLMRoberta even if a local path is configured.

  #102 opened 3 years ago by yym6472
  2

• Do openKiwi have confident score?

  #103 opened 3 years ago by baoloc97
  1

• Error Pre-Training Predictor: "model -> encoder -> encode_source extra fields not permitted (type=value_error.extra)"

  #104 opened 3 years ago by Qbop981001
  1

• I suppose that the code comment should be remove.

  #96 opened 3 years ago by zh25714
  2

• The prediction process is not complete by Predictor Estimator.

  #101 opened 3 years ago by Balakiranred
  5

• Poor results when training Estimator with parallel data and TER scores

  #92 opened 4 years ago by lluisg
  1

• Error at Predictor Training: "Predictor is not a subclass of QESystem"

  #97 opened 4 years ago by tervingo
  2

• TypeError: cannot unpack non-iterable NoneType object

  #100 opened 3 years ago by leileilin
  1

• OSError: Can't load weights for 'xlm-roberta-base'.

  #98 opened 3 years ago by genbei
  16

• some problems about data without alignments

  #95 opened 4 years ago by leileilin
  11

• some confusions

  #94 opened 4 years ago by leileilin
  0

• openkiwi version

  #93 opened 4 years ago by leileilin
  2

• training failed when only predicting source tags

  #91 opened 4 years ago by iamhere1
  10

• Add transformer encoder fine-tuning with Adapters

  #88 opened 4 years ago by daandouwe
  0

• Add calibration and active learning with Baal

  #87 opened 4 years ago by daandouwe
  0

• Add Streamlit demo

  #71 opened 4 years ago by daandouwe
  1

• Relative links break when displayed in docs

  #84 opened 4 years ago by captainvera
  0

• Error happened when running demo

  #81 opened 4 years ago by iamhere1
  7

• Unable to start openkiwi

  #79 opened 4 years ago by knikolaou48
  1

• Evaluate pipeline cannot deal with `verbose` and `quiet` arguments

  #82 opened 4 years ago by daandouwe
  0

• Hyperparamete tuning using skorch

  #67 opened 4 years ago by danparamov
  2

• Add alignment prediction with SimAlign

  #77 opened 4 years ago by daandouwe
  0

• Hi, I am trying to load the model once and then predict to stdout as the input is fed to stdin. I cannot find any module in the python package to do this.

  #54 opened 4 years ago by Bhavani01
  4

• TypeError: metaclass conflict: the metaclass of a derived class must be a (non-strict) subclass of the metaclasses of all its bases

  #75 opened 4 years ago by Shannen3206
  4

• I have use quickstart .

  #66 opened 4 years ago by cyjack
  2

• Pre-trained models for v. 2.0.0

  #69 opened 4 years ago by andrasaponyi
  0

• Replacing the predictor by BERT or XLM to implement PREDEST-BERT and PREDEST-XLM?

  #65 opened 4 years ago by AriaChen
  7

• how fast is the predictor training supposed to be?

  #55 opened 4 years ago by BigBorg
  2

• Different servers to load the trained PredEst model issue

  #53 opened 4 years ago by HanchengDeng
  4

• How can I continue my predictor training when interrupted?

  #63 opened 4 years ago by hcdeng6
  2

• Vocabulary size for predictor training

  #64 opened 4 years ago by andrasaponyi
  0

• Can i use openkiwi to do data cleaning?

  #56 opened 4 years ago by BigBorg
  2

• Error in predicting when using GPU for predictor-estimator

  #60 opened 4 years ago by prabhakar267
  9

• Documentation: Can't find yaml configuration files for QUETCH

  #62 opened 4 years ago by warlock2k
  6

• Question about data

  #57 opened 5 years ago by mzeidhassan
  4

• Unable to predict gap tags

  #59 opened 5 years ago by zouharvi
  5

• array must not contain infs or NaNs

  #47 opened 5 years ago by BigBorg
  7

• GPU training crashes

  #43 opened 5 years ago by Bhavani01
  36

• Is it possible to train with just src, mt, ter?

  #46 opened 5 years ago by jimbogill
  3