Do I have to train 2 model for 2 direction of translate En - De and De - En?

[duynx (Duy Nguyen)]

Hello, I have a model to translate from English to German. When I want to translate from German to English.
Do I have to train one more model in the opposite direction? or
Are Opennmt have some method to use 1 model for 2 direction of translate?
Thank you!

[guillaumekln (Guillaume Klein)]

Do I have to train one more model in the opposite direction?

That is what is usually done.

Are Opennmt have some method to use 1 model for 2 direction of translate?

This topic could be useful:

Expanding the model to two-directional and/or Multi-Language Support

Hi, I have come up an English to Spanish (Domain Adapted) model that does well. This involved me coming up with parallel data, setting up lot of pre-processing steps, developing the base model and then fine-tuning on domain data. I’m wondering how can I do the following with OpenNMT-tf repo - Create a two directional model - one that performs both English --> Spanish and Spanish --> English. Is this even possible or I have to follow the same process steps twice ending up with two models one …