fairseq

fairseq copied to clipboard

## ❓ Questions and Help ### Before asking: 1. search the issues. 2. search the docs. #### What is your question? The NLLB-200 translation model is benefiting humanity. Are there...

youningnihaobang

Hello, I am interested on running MMS ASR models in language-agnostic mode (i.e., input test audio's language ID is unknown). The similar results have been reported in the MMS paper....

khassanoff

I am trying to run trainbart.sh script. the script contains the following information: `TOTAL_NUM_UPDATES=20000 WARMUP_UPDATES=500 LR=3e-05 MAX_TOKENS=1024 UPDATE_FREQ=16 BART_PATH=/home/mdrullah/MARMAID/fairseq/BERT_Large/bart.large/model.pt python train.py metaphor\ --restore-file $BART_PATH \ --max-tokens $MAX_TOKENS \ --task translation...

RifatUllah102

## 🐛 Bug ### To Reproduce Steps to reproduce the behavior (**always include the command you ran**): 1. Run cmd '....' 2. See error #### Code sample ### Expected behavior...

Myprojectjoy

I am using the fairseq library and I need to modify the source code to make it suitable for me. However, directly modifying the source code seems to be frowned...

lazysleepypanda

Hi, seems like SpeechMatrix has some missing files (#4889), and training instructions for Textless S2ST on Real Data haven't been added yet. Is there a plan for an update? Thanks

tjysdsg

# Before submitting - [ ] Was this discussed/approved via a Github issue? (no need for typos, doc improvements) - [x] Did you read the [contributor guideline](https://github.com/pytorch/fairseq/blob/main/CONTRIBUTING.md)? - [x] Did...

ftshijt

mms asr transcribe issues no solution

4

I am unable to run mms asr transcribe . I am using python3.11 and facing a range of issues from hypo.word not found, AttributeError: 'PosixPath' object has no attribute 'find'...

spanta28

NameError: name 'BertEmbeddings' is not defined

5

## 🐛 Bug Meet BertEmbedding not defined bug ### To Reproduce Use this colab to repro https://colab.research.google.com/drive/1jnF0XTF12cB1pV5OUH_-jscMIJQDmKa1?usp=sharing #### Code sample ### Expected behavior I need to run this code in...

Jiancong

In TransformerEncoderBase class, it's forward() function has a parameter 'src_tokens': tokens in the source language of shape `(batch, src_len)`. It's a tensor of indexes, suppoes that: [ [10, 52, 138,...

VTaPo