OpenNMT/OpenNMT-tf

Issues

• Feature Request: Support for TensorFlow 2.14 and Higher

  #1008 opened a year ago by ncicio
  1

• Multi-source NMT

  #1009 opened a year ago by ThuraAung1601
  0

• Wrong prediction length from exported OpenNMT-tf models with shared embeddings

  #1007 opened a year ago by dmar1n
  1

• tensorflow lite model example

  #1005 opened a year ago by flutter-painter
  8

• OpenNMT-tf onnx export ?

  #1006 opened a year ago by flutter-painter
  0

• Input size mismatch

  #1004 opened a year ago by pranjaliseth
  0

• word-level knowledge distillation

  #1001 opened 2 years ago by ZenBel
  0

• xla_ops failed when use multi gpu

  #1000 opened 2 years ago by wangshauitj
  0

• Quickstart example runs for a long time

  #998 opened 2 years ago by cspink
  5

• Make timeout value configurable while searching for an optimal batch size

  #996 opened 2 years ago by sandeepchittilla
  0

• "Horovod has been shut down" error when training is finished due to early stopping

  #985 opened 2 years ago by awery
  2

• Training fails with --mixed_precision and guided alignments

  #981 opened 2 years ago by Dagamies
  0

• ModuleNotFoundError: No module named 'lxml.etree'

  #980 opened 2 years ago by atanumandal0491
  1

• An input error occurred while expanding the dataset

  #979 opened 2 years ago by copperdong
  1

• Modify loss function or mask attention in this code

  #970 opened 2 years ago by madhurapande19
  1

• Parallel text inputter fails with compressed text files

  #973 opened 2 years ago by awery
  1

• Fine-tune GPT2 model from Transformers

  #965 opened 3 years ago by NeonBohdan
  4

• What material did you use to train your pretrained model?

  #964 opened 3 years ago by junyizhao04
  1

• SequenceClassifier doesn't seem to learn leaked target

  #957 opened 3 years ago by Ingvar-Y
  2

• cpp for tensorflow serving

  #956 opened 3 years ago by lyzKF
  2

• An issue with SequenceRecordInputter ?

  #952 opened 3 years ago by Ingvar-Y
  1

• Error when inferencing with gpt.

  #943 opened 3 years ago by songxxzp
  3

• RuntimeError: Model diverged with loss = NaN.

  #927 opened 3 years ago by rudyyin
  2

• Can I change opennmt-py model to opennmt-tf model?

  #937 opened 3 years ago by 520jefferson
  4

• Why BLEU score is computed during training and not during inference?

  #941 opened 3 years ago by code-rex1
  1

• Running sequence tagger using OpenNMT-tf

  #939 opened 3 years ago by faysalhossain2007
  2

• Decay scheduler based on validation score

  #933 opened 3 years ago by cccs-km
  4

• BPE support for transformer model

  #932 opened 3 years ago by nashid
  1

• Comparison of OpenNMT-tf and tensor2tensor

  #931 opened 3 years ago by nashid
  3

• TensorFlow Lite tests failing with TensorFlow 2.7+

  #903 opened 3 years ago by guillaumekln
  1

• After tflite invoke, the tflite model use a larger memory.

  #928 opened 3 years ago by SefaZeng
  5

• Why is the SelfAttentionEncoderlayer inconsistent with the paper?

  #930 opened 3 years ago by Wenjun-Peng
  2

• Multi GPU training fails with more than 2 A100 GPUs (RuntimeError: Model diverged with loss = NaN.)

  #918 opened 3 years ago by Dagamies
  20

• Fine tuning BERT models for NLP tasks

  #926 opened 3 years ago by cservan
  1

• RsqrtDecay fails when decay_params scale is configured

  #923 opened 3 years ago by Dagamies
  2

• Import error on mac M1

  #917 opened 3 years ago by vietph34
  1

• Is there a way to use dynamic decoding rather than fix output size for the tflite model?

  #916 opened 3 years ago by SefaZeng
  1

• Translator inference is repeating the word and including values when the .

  #912 opened 3 years ago by eduamf
  1

• Fail to load dynamic quantized tflite model

  #915 opened 3 years ago by SefaZeng
  5

• Tflite outputs a largers model file.

  #905 opened 3 years ago by SefaZeng
  9

• Tensorflow Lite failing with absolute position embedding?

  #904 opened 3 years ago by SefaZeng
  6

• On Quickstart Example: UnicodeDecodeError: 'utf-8' codec can't decode byte 0xc3 in position 0

  #886 opened 3 years ago by duvallj
  5

• RuntimeError: Model diverged with loss = NaN.

  #890 opened 3 years ago by FPBHW
  13

• Exception: SGD optimizer receives "beta1" which is not set

  #885 opened 3 years ago by FPBHW
  2

• SGD does not receive parameters

  #887 opened 3 years ago by FPBHW
  8

• CNN decoder

  #884 opened 3 years ago by AbdallahNasir
  2

• Output of inference with score formatted as a "byte string"

  #880 opened 4 years ago by kschweiger
  1

• Performance issues in the definition of _get_initial_state, opennmt/decoders/self_attention_decoder.py

  #870 opened 4 years ago by DLPerf
  2

• How to convert a fairseq model to OpenNMT type

  #871 opened 4 years ago by lishaojun412
  2

• CUDA_ERROR_ILLEGAL_ADDRESS: an illegal memory access

  #869 opened 4 years ago by Geremia
  2