Fairseq is a sequence modeling toolkit for training custom models for translation, summarization, and other text generation tasks. It provides reference implementations of various sequence-to-sequence models, including Long Short-Term Memory (LSTM) networks and a novel convolutional neural network (CNN) that can generate translations many times faster than comparable recurrent neural network (RNN) models.
Fairseq can train models that achieve state-of-the-art performance on machine translation and summarization tasks, and includes pre-trained models for several benchmark translation datasets.
Fairseq also features multi-GPU training on one or across multiple machines, and lightning fast beam search generation on both CPU and GGPU.
git clone https://github.com/pytorch/fairseq.git cd fairseq pip install -r requirements.txt python setup.py build develop
Download a pre-trained model to familarize yourself with fairseq-py.
Train a new model.