OpenNMT: Open Source Neural Machine Translation in Torch

#1

Post by sahin-boydas » Tue, Jul 31, 2018, 9:08 AM UTC

OpenNMT: Open Source Neural Machine Translation in Torch
github.com

Re: OpenNMT: Open Source Neural Machine Translation in Torch

#2

Post by nl » Thu, Aug 02, 2018, 11:59 AM UTC

Don't use this. (Lua)Torch is obsolete.

Instead, if you want OpenNMT specifically for some reason then you should use either https://github.com/OpenNMT/OpenNMT-tf (TensorFlow) or https://github.com/OpenNMT/OpenNMT-py (PyTorch).

I'm not sure what the best open source neural translation model is, but Google's Transformer models are very strong (and can be trained on TPUs, which is cheap(er) and fast): https://github.com/tensorflow/tensor2tensor/#translation

Re: OpenNMT: Open Source Neural Machine Translation in Torch

#3

Post by valedra » Thu, Aug 02, 2018, 3:38 PM UTC

nl wrote: post #2

Don't use this. (Lua)Torch is obsolete. Instead, if you want OpenNMT specifically for some reason then you should use either https://github.com/OpenNMT/OpenNMT-tf (TensorFlow) or https://github.com/OpenNMT/OpenNMT-py (PyTorch). I'm not sure what the best open source neural translation model is, but Google's Transformer models are very strong (and can be trained on TPUs, which is cheap(er) and fast): https://github.co…

I would choose the framework according to your goals - OpenNMT-py is very research-oriented and hackable. It supports Transformer, copy-attention, image/speech/text2text and more. If you are more production-focused, maybe MarianNMT or OpenNMT-tf are for you.