fairseq

fairseq copied to clipboard

#### Code this is a MRE: https://colab.research.google.com/drive/1twRnR4tJxhyJneJ552yqRli0fIKV5idK?usp=sharing `!pip install fairseq` and `!git clone https://github.com/facebookresearch/fairseq` `%cd fairseq` `!pip install -e .` yields ``` Collecting fairseq Using cached fairseq-0.12.2.tar.gz (9.6 MB) Installing...

cgiturtle

Converted via hf. Is it possible to convert the model to gguf and quantize it? ## input: #### python convert_hf_to_gguf.py ../nllb-200-tf --outfile nnlb-200.gguf --outtype f16 ## output: #### ERROR:hf-to-gguf:Model M2M100ForConditionalGeneration...

Dovegs

## 🐛 Bug WeightsUnpickler error: Unsupported global: GLOBAL numpy._core.multiarray._reconstruct was not an allowed global by default. Please use `torch.serialization.add_safe_globals([_reconstruct])` or the `torch.serialization.safe.globals([_reconstruct])` context manager to allowlist this global if you...

sa-paul

## 📚 Documentation # Issue Description Hi, I am providing an example of how to enable Monte Carlo Dropout during inference to improve accessibility and usability. FairSeq includes an implementation...

Kensuke-Mitsuzawa

clarification on additional_special_tokens

2

Is there a map telling me which of these special tokens map to which language? "additional_special_tokens": [ --   | "ace_Arab",   | "ace_Latn",   | "acm_Arab",   | "acq_Arab",   | "aeb_Arab",   | "afr_Latn",  ...

hwang136

roberta-base wrong number of parameters

1

## ❓ Questions and Help #### What is your question? When I followed the pre-training instructions（[examples/roberta/README.pretraining.md](https://github.com/facebookresearch/fairseq/blob/main/examples/roberta/README.pretraining.md)） and pre-trained roberta from scratch using roberta-base, I found that the number of parameters...

mumu12641

Gradio

2

## 🐛 Bug I was trying to use the mms using an api and the stt functionality doesn't work saying "Loaded as API: https://mms-meta-mms.hf.space ✔" followed by "An error occurred:...

Eyobs-droid

FSDP fails to work together with activation checkpointing under `translation_multi_simple_epoch` training transformer from scratch

5

## 🐛 Bug Got RuntimeError when training transformer from scratch under `translation_multi_simple_epoch` task with fully sharded data parallel (FSDP). ### To Reproduce Steps to reproduce the behavior (**always include the...

thpun

Speech recognition reproducibility

15

Hi, I am having trouble reproducing the speech recognition results. With the default settings, the model stagnates at 25% train accuracy. By employing a different optimizer, increasing the batch size...

Bobrosoft98

## ❓ Questions and Help I need language code or id of Saudi Arabia. I look at the HuggingFace website. But I don't understand which one it is. Where can...

Cafarli