facebookresearch/fairseq

Issues

• Failed to build fairseq ERROR: Could not build wheels for fairseq, which is required to install pyproject.toml-based projects

  #5472 opened a month ago by Yogier
  1

• Inference on MoE models

  #5497 opened 15 days ago by meenakshi-mittal
  0

• Add `--validate-after-epochs` training flag

  #5496 opened 16 days ago by mcognetta
  0

• src_seq_length fixed length error when training MoE model

  #5495 opened 16 days ago by meenakshi-mittal
  0

• Add Saraiki language

  #5494 opened 19 days ago by mpQadir
  0

• Request for Inclusion of Sindhi Language in Multilingual Model for Speech Recognition

  #5493 opened 19 days ago by buzair6
  0

• Training a model with a custom FairseqDataset implementation

  #5492 opened 25 days ago by sidharthrajaram
  0

• object of type 'NoneType' has no len()

  #5491 opened a month ago by nitinyadav250
  1

• Use symbolic link for saving best/last checkpoints

  #5490 opened a month ago by ZeroRin
  0

• How can I get WER (train/valid) for the audio_finetuning task with CTC?

  #5489 opened a month ago by Addison-Weatherhead
  0

• 'TranslationTask' object has no attribute 'args'

  #5488 opened a month ago by whiteSteelRain
  0

• Addition dot in translated text

  #5487 opened a month ago by gosha70
  0

• Mandarin and Cantonese language model not found

  #5483 opened a month ago by imMonicaShaw
  0

• This model has order 20 but KenLM was compiled to support up to 6

  #5482 opened a month ago by imMonicaShaw
  0

• cannot translate the whole paragraph/sentences

  #5478 opened a month ago by junxu-ai
  1

• Multi-modal data2vec (joint training)

  #5481 opened a month ago by netw0rkf10w
  0

• decoding in Hubert: RecursionError: maximum recursion depth exceeded in comparison

  #5480 opened a month ago by Charlottehoo
  0

• RuntimeError: Expected 2D (unbatched) or 3D (batched) input to conv1d, but got input of size: [1, 1, 191, 80]

  #5479 opened a month ago by Maverick7t
  0

• How MMS uses microphones in real-time

  #5477 opened a month ago by monkeycc
  0

• Removing logs of Fairseq

  #5476 opened a month ago by Akhil-VSSG
  0

• Empty 'args' value in Neural Language Modeling "Training a transformer language model with the CLI tools" example model

  #5467 opened 2 months ago by lancioni
  2

• AttributeError when trying to train model in Fairseq

  #5473 opened a month ago by Maverick7t
  0

• the new LID Model lid218e.bin doesn't detect Chinese/Japanese correctly.

  #5438 opened 3 months ago by jingkang99
  1

• How to use fairseq.models.transformer.TransformerModel.from_pretrained on Multilingual translation model

  #5470 opened 2 months ago by JoeyandLucy
  0

• Arabizi/ Franco-Arabic text translated to English

  #5469 opened 2 months ago by omnipervius
  0

• fairseq package isn't working with python 3.11.8 as it supposed to be.

  #5457 opened 2 months ago by K-S-King
  1

• Question about data preparation with speech data alignment in speech matrix dataset

  #5466 opened 2 months ago by tony-sensei
  0

• calculation of the perplexity score

  #5464 opened 2 months ago by deven367
  0

• ModuleNotFoundError: No module named 'fairseq.criterions.'

  #5454 opened 2 months ago by libeineu
  2

• Why this project alway force me to install CUDA12.1 ???

  #5462 opened 2 months ago by piwawa
  0

• generated an exception: Failed to decode audio While running asr_prep_json.py on custom dataset

  #5461 opened 2 months ago by 3sakshij
  0

• How did you train the k-means clustering model on the HuBERT model?

  #5460 opened 2 months ago by Remaxic
  0

• Need help in running prepare_text.sh for Wav2Vec2-U

  #5459 opened 2 months ago by Nemesis-19
  0

• zero-shot prediction on MMPT VidCLIP model warning `Some weights of the model checkpoint at bert-base-uncased were not used when initializing MMBertForEncoder`

  #5456 opened 2 months ago by november07
  1

• FileNot Found error

  #5455 opened 2 months ago by Maverick7t
  0

• [Hubert] Use different kmeans models for train and valid dataset?

  #5453 opened 2 months ago by amadeusuzx
  0

• MMS 下载预训练模型MMS-1B:L1107，想测试一下安多藏语

  #5451 opened 2 months ago by didi222-lqq
  0

• assert step < max_len, f"{step} < {max_len}" AssertionError: 60 < 60

  #5452 opened 2 months ago by sevensix617
  0

• Importing `hydra.experimental` results with an ImportError

  #5450 opened 2 months ago by SagiPolaczek
  0

• AssertionError: Sentences lengths should not exceed max_tokens=400000

  #5447 opened 3 months ago by BirdiD
  1

• Installation Error: M1 Mac

  #5448 opened 3 months ago by vandit2209
  1

• Any plan to support TTS for cmn?

  #5446 opened 3 months ago by JustinQiu
  0

• raise FileNotFoundError("Dict not found: {}".format(args.data))

  #5445 opened 3 months ago by guo-chuanchuan
  0

• Roberta base for Custom named-entity recognization

  #5444 opened 3 months ago by Project-NSE
  0

• issue numpy when try to TTS method

  #5443 opened 3 months ago by changizyv
  0

• Wrong results when running inference with pytorch and MPS backend

  #5442 opened 3 months ago by tonytorm
  1

• Fairseq inference with Tauri frontend

  #5441 opened 3 months ago by kabyanil
  0

• nllb 3.3B translate from Chinese to Korean got: , , , , , , , , , , , , , , ,

  #5440 opened 3 months ago by micronetboy
  0

• nllb 3.3B translate from Chinese to Japanese got: () () () () () () () () () () () () ()

  #5439 opened 3 months ago by micronetboy
  0

• Why the nllb 3.3B will still occupy 5.7GB memory while the model has loaded to GPU and occupied 13.17GB GPU-memory？

  #5437 opened 4 months ago by micronetboy
  0