fairseq

fairseq copied to clipboard

'dict' object has no attribute 'task' while decoding the language model

2

When i'm trying to run the inference of the finetuned model along with transformer based language model which i'v trained i'm getting this error: ``` INFO:__main__:| decoding with criterion ctc...

mukherjeesougata-eros

I downloaded the model and everything related but getting request error from https://dl.fbaipublicfiles.com/fairseq/gpt2_bpe/encoder.json

3

## 🐛 Bug I have all the files locally including the encoder but from time to time I get requests error from https://dl.fbaipublicfiles.com/fairseq/gpt2_bpe/encoder.json website which indicates the from_pretrained function tries...

KemalDrop

Typo fix for fast_noisy_channel README.md

3

Small fix to fastbpe preprocessing script in examples/fast_noisy_channel/README.md.

alexandremuzio

fix/typo/Readme

3

# 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)? - [ ]...

YaYaB

hello everyone: I write a new task named "speech_translation" and it is trained by CTC loss. I want to use the file "examples/speech_recognition/new/infer.py" to evaluate its performance. ``` python3 /fairseq/examples/speech_recognition/new/infer.py...

fengpeng-yue

How to use the past, future and self targets for text generation?

3

#### What is your question? I noticed that new FairSeq models support past, future and self targets for text generation tasks. However, I have been trying to figure out a...

AhtiAhde

nit

1

# 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)? - [...

dianaml0

Hi，I found that when the bpe used when data set was prepared(such as: bash prepare-iwslt14.sh).How can I add the bpe-dropout while training the model on each batch separately?

lzp-man

## 🐛 Bug When training an spm model with byte fallback, the decoded output in fairseq doesn't replace the bytes with the appropriate character. ### To Reproduce Train spm model...

erip

ShunchiZhang