Translate - a PyTorch Language Library

Translate is a library for machine translation written in PyTorch. It provides training for sequence-to-sequence models. Translate relies on fairseq, a general sequence-to-sequence library, which means that models implemented in both Translate and Fairseq can be trained. Translate also provides the ability to export some models to Caffe2 graphs via ONNX and to load and run these models from C++ for production purposes. Currently, we export components (encoder, decoder) to Caffe2 separately and beam search is implemented in C++. In the near future, we will be able to export the beam search as well. We also plan to add export support to more models.

Quickstart

If you are just interested in training/evaluating MT models, and not in exporting the models to Caffe2 via ONNX, you can install Translate for Python 3 by following these few steps:

Install pytorch
Install fairseq
Clone this repository git clone https://github.com/pytorch/translate.git pytorch-translate && cd pytorch-translate
Run python setup.py install

Provided you have CUDA installed you should be good to go.

Requirements and Full Installation

Translate Requires:

A Linux operating system with a CUDA compatible card
GNU C++ compiler version 4.9.2 and above
A CUDA installation. We recommend CUDA 8.0 or CUDA 9.0

Use Our Docker Image:

Install Docker and nvidia-docker, then run

sudo docker pull pytorch/translate
sudo nvidia-docker run -i -t --rm pytorch/translate /bin/bash
. ~/miniconda/bin/activate
cd ~/translate

You should now be able to run the sample commands in the Usage Examples section below. You can also see the available image versions under https://hub.docker.com/r/pytorch/translate/tags/.

Install Translate from Source:

These instructions were mainly tested on Ubuntu 16.04.5 LTS (Xenial Xerus) with a Tesla M60 card and a CUDA 9 installation. We highly encourage you to report an issue if you are unable to install this project for your specific configuration.

If you don't already have an existing Anaconda environment with Python 3.6, you can install one via Miniconda3:

wget https://repo.continuum.io/miniconda/Miniconda3-latest-Linux-x86_64.sh -O miniconda.sh
chmod +x miniconda.sh
./miniconda.sh -b -p ~/miniconda
rm miniconda.sh
. ~/miniconda/bin/activate

Clone the Translate repo:

git clone https://github.com/pytorch/translate.git
pushd translate

Install the PyTorch conda package:

# Set to 8 or 9 depending on your CUDA version.
TMP_CUDA_VERSION="9"

# Uninstall previous versions of PyTorch. Doing this twice is intentional.
# Error messages about torch not being installed are benign.
pip uninstall -y torch
pip uninstall -y torch

# This may not be necessary if you already have the latest cuDNN library.
conda install -y cudnn

# Add LAPACK support for the GPU.
conda install -y -c pytorch "magma-cuda${TMP_CUDA_VERSION}0"

# Install the combined PyTorch nightly conda package.
conda install pytorch-nightly cudatoolkit=${TMP_CUDA_VERSION}.0 -c pytorch

# Install NCCL2.
wget "https://s3.amazonaws.com/pytorch/nccl_2.1.15-1%2Bcuda${TMP_CUDA_VERSION}.0_x86_64.txz"
TMP_NCCL_VERSION="nccl_2.1.15-1+cuda${TMP_CUDA_VERSION}.0_x86_64"
tar -xvf "${TMP_NCCL_VERSION}.txz"
rm "${TMP_NCCL_VERSION}.txz"

# Set some environmental variables needed to link libraries correctly.
export CONDA_PATH="$(dirname $(which conda))/.."
export NCCL_ROOT_DIR="$(pwd)/${TMP_NCCL_VERSION}"
export LD_LIBRARY_PATH="${CONDA_PATH}/lib:${NCCL_ROOT_DIR}/lib:${LD_LIBRARY_PATH}"

Install ONNX:

git clone --recursive https://github.com/onnx/onnx.git
yes | pip install ./onnx 2>&1 | tee ONNX_OUT

If you get a Protobuf compiler not found error, you need to install it:

conda install -c anaconda protobuf

Then, try to install ONNX again:

yes | pip install ./onnx 2>&1 | tee ONNX_OUT

Build Translate:

pip uninstall -y pytorch-translate
python3 setup.py build develop

Now you should be able to run the example scripts below!

Usage Examples

Note: the example commands given assume that you are the root of the cloned GitHub repository or that you're in the translate directory of the Docker or Amazon image. You may also need to make sure you have the Anaconda environment activated.

Training

We provide an example script to train a model for the IWSLT 2014 German-English task. We used this command to obtain a pretrained model:

bash pytorch_translate/examples/train_iwslt14.sh

The pretrained model actually contains two checkpoints that correspond to training twice with random initialization of the parameters. This is useful to obtain ensembles. This dataset is relatively small (~160K sentence pairs), so training will complete in a few hours on a single GPU.

Training with tensorboard visualization

We provide support for visualizing training stats with tensorboard. As a dependency, you will need tensorboard_logger installed.

pip install tensorboard_logger

Please also make sure that tensorboard is installed. It also comes with tensorflow installation.

You can use the above example script to train with tensorboard, but need to change line 10 from :

CUDA_VISIBLE_DEVICES=0 python3 pytorch_translate/train.py

CUDA_VISIBLE_DEVICES=0 python3 pytorch_translate/train_with_tensorboard.py

The event log directory for tensorboard can be specified by option --tensorboard_dir with a default value: run-1234. This directory is appended to your --save_dir argument.

For example in the above script, you can visualize with:

tensorboard --logdir checkpoints/runs/run-1234

Multiple runs can be compared by specifying different --tensorboard_dir. i.e. run-1234 and run-2345. Then

tensorboard --logdir checkpoints/runs

can visualize stats from both runs.

Pretrained Model

A pretrained model for IWSLT 2014 can be evaluated by running the example script:

bash pytorch_translate/examples/generate_iwslt14.sh

Note the improvement in performance when using an ensemble of size 2 instead of a single model.

Exporting a Model with ONNX

We provide an example script to export a PyTorch model to a Caffe2 graph via ONNX:

bash pytorch_translate/examples/export_iwslt14.sh

This will output two files, encoder.pb and decoder.pb, that correspond to the computation of the encoder and one step of the decoder. The example exports a single checkpoint (--checkpoint model/averaged_checkpoint_best_0.pt but is also possible to export an ensemble (--checkpoint model/averaged_checkpoint_best_0.pt --checkpoint model/averaged_checkpoint_best_1.pt). Note that during export, you can also control a few hyperparameters such as beam search size, word and UNK rewards.

Using the Model

To use the sample exported Caffe2 model to translate sentences, run:

echo "hallo welt" | bash pytorch_translate/examples/translate_iwslt14.sh

Note that the model takes in BPE inputs, so some input words need to be split into multiple tokens. For instance, "hineinstopfen" is represented as "hinein@@ stop@@ fen".

PyTorch Translate Research

We welcome you to explore the models we have in the pytorch_translate/research folder. If you use them and encounter any errors, please paste logs and a command that we can use to reproduce the error. Feel free to contribute any bugfixes or report your experience, but keep in mind that these models are a work in progress and thus are currently unsupported.

Join the Translate Community

We welcome contributions! See the CONTRIBUTING.md file for how to help out.

License

Translate is BSD-licensed, as found in the LICENSE file.

jinfengr/translate