facebookresearch fairseq issues

facebookresearch / fairseq

Facebook AI Research Sequence-to-Sequence Toolkit written in Python.

MIT License

30.58k stars 6.41k forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

score calculation in translation_moe

#5566 zhmingzhaung opened 2 days ago
1
How do I test it after training?wav2vec2

#5565 lucky-asia opened 1 week ago
0
How to use fairseq-hydra-train to finetune Wav2Vec on Intel accelerators?

#5564 vijethk-intel opened 2 weeks ago
0
How to generate hypothese with fconv for DrNMT? (ende)

#5563 123THU opened 2 weeks ago
0
question about mms tts text input

#5562 zhoujxwilliam opened 2 weeks ago
1
long-standing Bug in Adafactor optimizer if beta1 > 0

#5561 dxqbYD opened 3 weeks ago
0
Do I need to crop long audio for inference based on pretrained models？

#5560 xuduo18311199384 opened 3 weeks ago
1
clarification on additional_special_tokens

#5559 hwang136 opened 1 month ago
1
wav2vec2-object of type 'NoneType' has no len()

#5558 Linx3f opened 1 month ago
0
Include Kirundi (Rundi - run_Latn) language in the No Language Left Behind (NLLB) project

#5557 labKnowledge opened 1 month ago
1
Can't find fairseq/version.txt during pip installation

#5556 GryffindorLi closed 1 week ago
2
Add an `auto_expand` option to `SinusoidalPositionalEmbedding`

#5555 MaigoAkisame closed 1 month ago
0
How to solve the problem of continuing pre-training the XLSR-53 2B model

#5554 Learneducn opened 1 month ago
0
When I use 8 A40 images to continue pre-training the XLSR-53 2B pre-training model, I always get an error message: Expected is_sm80 to be true, but got false. (could this error message be improved? If so, please report an enhancement request to pytorch) How can I solve this problem?

#5553 Learneducn opened 1 month ago
1
Error when installing fairseq

#5552 ChristopheYe opened 1 month ago
4
slk missing

#5551 userghubhub opened 1 month ago
0
Skip some tokens

#5550 ekmekovski opened 1 month ago
0
NLLB is unable to translate into a complete long sentence in Chinese.

#5549 logicvv opened 1 month ago
1
Delete examples/speech_synthesis directory

#5548 wailin247 opened 1 month ago
1
update paper link + bug fix

#5547 brianyan918 closed 1 month ago
0
Update readme with arxiv link

#5546 brianyan918 closed 1 month ago
0
Add LID rerank for MMS

#5545 brianyan918 closed 1 month ago
0
Clean scripts and add README

#5544 brianyan918 closed 1 month ago
3
ValueError: offset must be non-negative and no greater than buffer length

#5543 LiYixuan727 opened 2 months ago
3
about load dgslm

#5542 VJJJJJJ1 closed 2 months ago
0
non-autoregressive transformer training scripts error

#5541 ChloeKiwi opened 2 months ago
0
Is there a C++ version of Fairseq based on libtorch

#5540 EbanShen opened 2 months ago
0
Switch version and make change

#5539 henrycharlesworth closed 2 months ago
1
Has anyone got the MMPT example to work?

#5538 qingy1337 opened 2 months ago
1
Inferring the decoder and encoder of a transformer model in a streaming manner

#5537 pengpengtao opened 2 months ago
0
Some questions about difference fairseq checkpoint_wrapper and fairscale checkpoint_wrapper

#5536 ZhikangNiu opened 3 months ago
0
unkpen does not work

#5535 LiYixuan727 opened 3 months ago
0
pretraining Hubert on Music Data

#5534 Bloos opened 3 months ago
0
Wav2Vec2 Pretraining

#5533 rajeevbaalwan opened 3 months ago
3
Overflow issue with Fairseq Preprocess for large datasets

#5532 henrycharlesworth opened 3 months ago
0
The loss value is extremely high when fine-tuning the xm_transformer_unity model. Is there a wrong step?

#5531 wanhhe opened 3 months ago
0
using restore-file option, loss became large

#5530 TimothyChen225 closed 4 months ago
0
MMS Zero-shot release

#5529 vineelpratap closed 4 months ago
0
Prebuilt wheels provided via 3rd party repository

#5528 MiroPsota opened 4 months ago
0
Bug in TokenBlockDataset causing potential CUDA OOM due to incorrect size casting

#5527 YuvalRingel opened 4 months ago
0
Pretrain Hubert on english and chinese speech dataset.

#5526 shihuai closed 2 days ago
8
Apex installation instructions update

#5525 mattf1n opened 4 months ago
1
Is NCCL really necessary for single-GPU training?

#5524 Geremia opened 4 months ago
0
replaced depracated np.float with np.float64

#5523 surprisedPikachu007 opened 4 months ago
0
task: invalid choice: 'speech_dlm_task'

#5522 HelenJi-maker opened 4 months ago
0
NLLB-200 1.3B token dimension compatibility

#5521 leqij opened 4 months ago
0
[MMS-TTS] how to mark the speaker‘s character tones, can't find any documents

#5520 shuxiang opened 4 months ago
0
Is there any plans to support torch.export?

#5519 mhs4670go opened 4 months ago
0
No such file or directory: 'fairseq\\version.txt

#5518 Strilin opened 4 months ago
6
Unable to pip install fairseq.

#5517 Valerie9696 closed 4 months ago
1