Concept information
Terme préférentiel
seq2seq neural machine translation
Définition
- Approach to machine translation that uses deep learning neural networks to directly translate sequences of words from one language to another. It typically consists of a two-part model: an encoder that reads in the input sentence and generates a thought vector representation of its meaning, and a decoder that takes the thought vector and produces the output sentence word by word.
Concept générique
Traductions
URI
http://data.loterre.fr/ark:/67375/8LP-GRGH8VG1-G
{{label}}
{{#each values }} {{! loop through ConceptPropertyValue objects }}
{{#if prefLabel }}
{{/if}}
{{/each}}
{{#if notation }}{{ notation }} {{/if}}{{ prefLabel }}
{{#ifDifferentLabelLang lang }} ({{ lang }}){{/ifDifferentLabelLang}}
{{#if vocabName }}
{{ vocabName }}
{{/if}}