Fairseq SimpleLSTM tutorial

Resulting files from running the tutorial at the Fairseq SimpleLSTM tutorial. The scripts added model only needs to be located at this directory and have fairseq installed to run. No need to edit the fairseq installation.

This should work as a template for other fairseq projects with plugins.

Getting Started

In order to get the model training and generation tasks running with fairseq we do as follows:

Obtaining the Pre-processed Data

Install fairseq on another directory (not here!) as follows:

git clone https://github.com/pytorch/fairseq
cd fairseq
pip install --editable ./

# on MacOS:
# CFLAGS="-stdlib=libc++" pip install --editable ./

# to install the latest stable release (0.10.x)
# pip install fairseq

After that, while on the new fairseq/ directory run the following in order to create the preprocessed training data:

cd examples/translation/
bash prepare-iwslt14.sh
cd ../..
TEXT=examples/translation/iwslt14.tokenized.de-en
fairseq-preprocess --source-lang de --target-lang en \
    --trainpref $TEXT/train --validpref $TEXT/valid --testpref $TEXT/test \
    --destdir data-bin/iwslt14.tokenized.de-en

After that, run the following line to move newly obtained data to this directory:

cp -rf ./data-bin/iwslt14.tokenized.de-en $PATH_TO_LSTM_TUTORIAL/data-bin/

Running training

From the project root run:

bash ./scripts/train_model.sh

Running generation