fairseq

fairseq copied to clipboard

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

hui-li-xf

mBART50 Translation/Fine Tuning with Many-to-One Model not working

19

Hello, I think there is something wrong with the arch. If I get the args for the mbart50.ft.n1 model it says "arch": "denoising_large". But denoising_large is not available in Fairseq...

sergej-schultenkaemper

Hello team, I'm working on a speaker profiling project that depends on fairseq. However, I’m consistently facing build failures related to libbleu.cpp and a missing standard header file (stddef.h) during...

pxD1311

## ❓ Questions and Help What is your question? Is there a way to use RoBERTa with sequences longer than 512 tokens (e.g., 2,000+)? I know RoBERTa was pretrained with...

VicenteLavagnino