Neural Machine Translation from English to German: Deployment Workflow

KCH Workflows
1

This is the deployment workflow of the encoder-decoder neural architecture for the Neural machine Translation model. It includes reading the encoder and decoder networks from tensorFlow files, applying them to English sentences and create the German character sequence as output.

This is a companion discussion topic for the original entry at https://kni.me/w/NtfdfxkUvjdSqZYp
2

Hi, Iʻm novice with Knime and NMT.
What is the difference between the Training workflow, the deployment workflow and the Neural Machine Translation English to German workflow?

Also, I would like to use a word-level encoder-decoder network of LSTMs instead of character-level. Is there any Knime documentation for that?