Skip to content
This repository has been archived by the owner on Oct 26, 2022. It is now read-only.

facebookresearch/fairseq-lua

Repository files navigation

Introduction

Note: there is now a PyTorch version of this toolkit (fairseq-py) and new development efforts will focus on it. The Lua version is preserved here, but is provided without any support.

This is fairseq, a sequence-to-sequence learning toolkit for Torch from Facebook AI Research tailored to Neural Machine Translation (NMT). It implements the convolutional NMT models proposed in Convolutional Sequence to Sequence Learning and A Convolutional Encoder Model for Neural Machine Translation as well as a standard LSTM-based model. It features multi-GPU training on a single machine as well as fast beam search generation on both CPU and GPU. We provide pre-trained models for English to French, English to German and English to Romanian translation.

Model

Citation

If you use the code in your paper, then please cite it as:

@article{gehring2017convs2s,
  author          = {Gehring, Jonas and Auli, Michael and Grangier, David and Yarats, Denis and Dauphin, Yann N},
  title           = "{Convolutional Sequence to Sequence Learning}",
  journal         = {ArXiv e-prints},
  archivePrefix   = "arXiv",
  eprinttype      = {arxiv},
  eprint          = {1705.03122},
  primaryClass    = "cs.CL",
  keywords        = {Computer Science - Computation and Language},
  year            = 2017,
  month           = May,
}

and

@article{gehring2016convenc,
  author          = {Gehring, Jonas and Auli, Michael and Grangier, David and Dauphin, Yann N},
  title           = "{A Convolutional Encoder Model for Neural Machine Translation}",
  journal         = {ArXiv e-prints},
  archivePrefix   = "arXiv",
  eprinttype      = {arxiv},
  eprint          = {1611.02344},
  primaryClass    = "cs.CL",
  keywords        = {Computer Science - Computation and Language},
  year            = 2016,
  month           = Nov,
}

Requirements and Installation

  • A computer running macOS or Linux
  • For training new models, you'll also need a NVIDIA GPU and NCCL
  • A Torch installation. For maximum speed, we recommend using LuaJIT and Intel MKL.
  • A recent version nn. The minimum required version is from May 5th, 2017. A simple luarocks install nn is sufficient to update your locally installed version.

Install fairseq by cloning the GitHub repository and running

luarocks make rocks/fairseq-scm-1.rockspec

LuaRocks will fetch and build any additional dependencies that may be missing. In order to install the CPU-only version (which is only useful for translating new data with an existing model), do

luarocks make rocks/fairseq-cpu-scm-1.rockspec

The LuaRocks installation provides a command-line tool that includes the following functionality:

  • fairseq preprocess: Data pre-processing: build vocabularies and binarize training data
  • fairseq train: Train a new model on one or multiple GPUs
  • fairseq generate: Translate pre-processed data with a trained model
  • fairseq generate-lines: Translate raw text with a trained model
  • fairseq score: BLEU scoring of generated translations against reference translations
  • fairseq tofloat: Convert a trained model to a CPU model
  • fairseq optimize-fconv: Optimize a fully convolutional model for generation. This can also be achieved by passing the -fconvfast flag to the generation scripts.

Quick Start

Training a New Model

Data Pre-processing

The fairseq source distribution contains an example pre-processing script for the IWSLT14 German-English corpus. Pre-process and binarize the data as follows:

$ cd data/
$ bash prepare-iwslt14.sh
$ cd ..
$ TEXT=data/iwslt14.tokenized.de-en
$ fairseq preprocess -sourcelang de -targetlang en \
  -trainpref $TEXT/train -validpref $TEXT/valid -testpref $TEXT/test \
  -thresholdsrc 3 -thresholdtgt 3 -destdir data-bin/iwslt14.tokenized.de-en

This will write binarized data that can be used for model training to data-bin/iwslt14.tokenized.de-en.

Training

Use fairseq train to train a new model. Here a few example settings that work well for the IWSLT14 dataset:

# Standard bi-directional LSTM model
$ mkdir -p trainings/blstm
$ fairseq train -sourcelang de -targetlang en -datadir data-bin/iwslt14.tokenized.de-en \
  -model blstm -nhid 512 -dropout 0.2 -dropout_hid 0 -optim adam -lr 0.0003125 -savedir trainings/blstm

# Fully convolutional sequence-to-sequence model
$ mkdir -p trainings/fconv
$ fairseq train -sourcelang de -targetlang en -datadir data-bin/iwslt14.tokenized.de-en \
  -model fconv -nenclayer 4 -nlayer 3 -dropout 0.2 -optim nag -lr 0.25 -clip 0.1 \
  -momentum 0.99 -timeavg -bptt 0 -savedir trainings/fconv

# Convolutional encoder, LSTM decoder
$ mkdir -p trainings/convenc
$ fairseq train -sourcelang de -targetlang en -datadir data-bin/iwslt14.tokenized.de-en \
  -model conv -nenclayer 6 -dropout 0.2 -dropout_hid 0 -savedir trainings/convenc

By default, fairseq train will use all available GPUs on your machine. Use the CUDA_VISIBLE_DEVICES environment variable to select specific GPUs or -ngpus to change the number of GPU devices that will be used.

Generation

Once your model is trained, you can translate with it using fairseq generate (for binarized data) or fairseq generate-lines (for text). Here, we'll do it for a fully convolutional model:

# Optional: optimize for generation speed
$ fairseq optimize-fconv -input_model trainings/fconv/model_best.th7 -output_model trainings/fconv/model_best_opt.th7

# Translate some text
$ DATA=data-bin/iwslt14.tokenized.de-en
$ fairseq generate-lines -sourcedict $DATA/dict.de.th7 -targetdict $DATA/dict.en.th7 \
  -path trainings/fconv/model_best_opt.th7 -beam 10 -nbest 2
| [target] Dictionary: 24738 types
| [source] Dictionary: 35474 types
> eine sprache ist ausdruck des menschlichen geistes .
S	eine sprache ist ausdruck des menschlichen geistes .
O	eine sprache ist ausdruck des menschlichen geistes .
H	-0.23804219067097	a language is expression of human mind .
A	2 2 3 4 5 6 7 8 9
H	-0.23861141502857	a language is expression of the human mind .
A	2 2 3 4 5 7 6 7 9 9

CPU Generation

Use fairseq tofloat to convert a trained model to use CPU-only operations (this has to be done on a GPU machine):

# Optional: optimize for generation speed
$ fairseq optimize-fconv -input_model trainings/fconv/model_best.th7 -output_model trainings/fconv/model_best_opt.th7

# Convert to float
$ fairseq tofloat -input_model trainings/fconv/model_best_opt.th7 \
  -output_model trainings/fconv/model_best_opt-float.th7

# Translate some text
$ fairseq generate-lines -sourcedict $DATA/dict.de.th7 -targetdict $DATA/dict.en.th7 \
  -path trainings/fconv/model_best_opt-float.th7 -beam 10 -nbest 2
> eine sprache ist ausdruck des menschlichen geistes .
S	eine sprache ist ausdruck des menschlichen geistes .
O	eine sprache ist ausdruck des menschlichen geistes .
H	-0.2380430996418	a language is expression of human mind .
A	2 2 3 4 5 6 7 8 9
H	-0.23861189186573	a language is expression of the human mind .
A	2 2 3 4 5 7 6 7 9 9

Pre-trained Models

Generation with the binarized test sets can be run in batch mode as follows, e.g. for English-French on a GTX-1080ti:

$ fairseq generate -sourcelang en -targetlang fr -datadir data-bin/wmt14.en-fr -dataset newstest2014 \
  -path wmt14.en-fr.fconv-cuda/model.th7 -beam 5 -batchsize 128 | tee /tmp/gen.out
...
| Translated 3003 sentences (95451 tokens) in 136.3s (700.49 tokens/s)
| Timings: setup 0.1s (0.1%), encoder 1.9s (1.4%), decoder 108.9s (79.9%), search_results 0.0s (0.0%), search_prune 12.5s (9.2%)
| BLEU4 = 43.43, 68.2/49.2/37.4/28.8 (BP=0.996, ratio=1.004, sys_len=92087, ref_len=92448)

# Word-level BLEU scoring:
$ grep ^H /tmp/gen.out | cut -f3- | sed 's/@@ //g' > /tmp/gen.out.sys
$ grep ^T /tmp/gen.out | cut -f2- | sed 's/@@ //g' > /tmp/gen.out.ref
$ fairseq score -sys /tmp/gen.out.sys -ref /tmp/gen.out.ref
BLEU4 = 40.55, 67.6/46.5/34.0/25.3 (BP=1.000, ratio=0.998, sys_len=81369, ref_len=81194)

Join the fairseq community

License

fairseq is BSD-licensed. The license applies to the pre-trained models as well. We also provide an additional patent grant.

About

Facebook AI Research Sequence-to-Sequence Toolkit

Resources

License

Code of conduct

Security policy

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published