Zobrazit minimální záznam

 
dc.contributor.author Variš, Dušan
dc.date.accessioned 2022-03-17T16:24:10Z
dc.date.available 2022-03-17T16:24:10Z
dc.date.issued 2022-03-15
dc.identifier.uri http://hdl.handle.net/11234/1-4680
dc.description En-De translation models, exported via TensorFlow Serving, available in the Lindat translation service (https://lindat.mff.cuni.cz/services/translation/). The models were trained using the MCSQ social surveys dataset (available at https://repo.clarino.uib.no/xmlui/bitstream/handle/11509/142/mcsq_v3.zip). Their main use should be in-domain translation of social surveys. Models are compatible with Tensor2tensor version 1.6.6. For details about the model training (data, model hyper-parameters), please contact the archive maintainer. Evaluation on MCSQ test set (BLEU): en->de: 67.5 (train: genuine in-domain MCSQ data only) de->en: 75.0 (train: additional in-domain backtranslated MCSQ data) (Evaluated using multeval: https://github.com/jhclark/multeval)
dc.language.iso eng
dc.language.iso deu
dc.publisher Charles University, Faculty of Mathematics and Physics, Institute of Formal and Applied Linguistics (UFAL)
dc.relation info:eu-repo/grantAgreement/EC/H2020/823782
dc.rights Creative Commons - Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0)
dc.rights.uri http://creativecommons.org/licenses/by-nc-sa/4.0/
dc.subject machine translation
dc.subject transformer
dc.subject neural machine translation
dc.title MCSQ Translation Models (en-de) (v1.0)
dc.type toolService
metashare.ResourceInfo#ResourceComponentType#ToolServiceInfo.languageDependent true
metashare.ResourceInfo#ContentInfo.detailedType tool
dc.rights.label PUB
has.files yes
branding LINDAT / CLARIAH-CZ
contact.person Dušan Variš varis@ufal.mff.cuni.cz Charles University, Faculty of Mathematics and Physics, Institute of Formal and Applied Linguistics (UFAL)
sponsor European Union EC/H2020/823782 SSHOC - Social Sciences & Humanities Open Cloud euFunds info:eu-repo/grantAgreement/EC/H2020/823782
files.size 1399707302
files.count 2


 Soubory tohoto záznamu

Icon
Název
mcsq.de-en.zip
Velikost
663.55 MB
Formát
application/zip
Popis
German-to-English
MD5
b11a7d80a17127637e6862bc80e0f748
 Stáhnout soubor  Náhled
 Náhled souboru  
Icon
Název
mcsq.en-de.zip
Velikost
671.32 MB
Formát
application/zip
Popis
English-to-German
MD5
f38e69751aadf0ea687b8d574b59223e
 Stáhnout soubor  Náhled
 Náhled souboru  

Zobrazit minimální záznam