fairseq
fairseq copied to clipboard
facebookresearch
Facebook AI Research Sequence-to-Sequence Toolkit written in Python.
# 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)? - [...
``` #!/usr/bin/env bash fairseq-train \ data-bin/wmt14_en_de_distill \ --save-dir checkpoints \ --ddp-backend=legacy_ddp \ --task translation_lev \ --criterion nat_loss \ --arch nonautoregressive_transformer \ --noise full_mask \ --share-all-embeddings \ --optimizer adam --adam-betas '(0.9,0.98)'...
## 🚀 Feature Request A C++ version of Fairseq based on libtorch is required ### Motivation Pytorch and the entire python environment are too large after packaging, so use libtorch...
Hello, Is the dataset created by the authors of the paper "Scaling Technology to 1000+ languages" (https://arxiv.org/pdf/2305.13516.pdf) available publicly? Where can it be downloaded from? Thanks!
We test the performance of VideoClip through the video-text retrieval task on the COIN dataset, but the performance is much lower than the reported performance of VideoQA (26%
Is it possible to infer the model separately through encoder.onnx and decoder.onnx?
I am trying to run fairseq-train and I am passing --cpu but it's still giving error as following:- In Local -> AssertionError: Torch not compiled with CUDA enabled On GCP...
When I resume training from a saved checkpoint with 4 GPUs. In the main loop of training iterations (i.e., `for i, samples in enumerate(progress):`, I got an empty batch `samples=[{}]`...
# Before submitting - [x] 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)? ## What does this...
Metadata
Owner
Metadata
Facebook AI Research Sequence-to-Sequence Toolkit written in Python.