fairseq
fairseq copied to clipboard
facebookresearch
Facebook AI Research Sequence-to-Sequence Toolkit written in Python.
my fairseq version is 0.12.2, i use airseq-hydra-train for base model is ok when i use fairseq-hydra-train for base model with conformer backbone it shows like this: fairseq-hydra-train task.data=/path/to/LibriSpeech/ distributed_training.distributed_world_size=1...
## ❓ Questions and Help #### What is your question? I followed the fairseq data2vec 2.0 example https://github.com/facebookresearch/fairseq/tree/main/examples/data2vec for the speech recognition research and tried to reproduce the data2vec 2.0...
There is an incorrect behavior of LID model presented here (`lid218e.bin`). Particularly: `Как сообщает пресса` is identified as `rus_Cyrl` that is correct, but `КАК СООБЩАЕТ ПРЕССА` - as `eng_Latn`; `Добрый...
#### What is your question? How to use the language identification model trained on Flores-200 (mentioned in the NLLB paper) ? Model is presented in the repo but the utility...
Hello, Where can I find the list of languages LID model can identify? It seems that only FastText's list of 157 languages is published but it is an older version...
Norwegian is shown as fully supported on the chart in the original introduction at: https://ai.meta.com/blog/multilingual-model-speech-recognition/ However it is not listed among the supported languages in documentation. Which of the documents...
## ❓ Questions and Help #### What is your question? I am a beginner trying implement pretraining RoBERTa module by following the procedures present in [https://github.com/facebookresearch/fairseq/blob/main/examples/roberta/README.pretraining.md]. Dataset used is same...
## ❓ Questions and Help ### Before asking: 1. search the issues. 2. search the docs. #### What is your question? #### Code #### What have you tried? #### What's...
## 🐛 Bug ### To Reproduce Steps to reproduce the behavior (**always include the command you ran**): 1. Run finetune command in examples/roberta/commonsense_qa/README.md The training acc is always about 20%,...
# Before submitting - [√] Was this discussed/approved via a Github issue? (no need for typos, doc improvements) - [√] Did you read the [contributor guideline](https://github.com/pytorch/fairseq/blob/main/CONTRIBUTING.md)? - [√] Did you...
Metadata
Owner
Metadata
Facebook AI Research Sequence-to-Sequence Toolkit written in Python.