Sýna einfalda færslu atriðis

 
dc.contributor.author Jónsson, Haukur Páll
dc.contributor.author Loftsson, Hrafn
dc.contributor.author Steingrímsson, Steinþór
dc.date.accessioned 2020-06-24T14:40:46Z
dc.date.available 2020-06-24T14:40:46Z
dc.date.issued 2020-06-23
dc.identifier.uri http://hdl.handle.net/20.500.12537/46
dc.description Moses phrase-based statistical machine translation (Moses PBSMT) is a system which is used to develop and run machine translation models. It is distributed here as four packages: 1. Code from a github repository to train and run models. 2. Pretrained is-en system (Docker) 3. Pretrained en-is system (Docker) 4. Frontend to pre- and postprocess text for translation (Docker) The models here are not (exactly) the same as were used for human evaluation. These models have additionally been trained on open dictionaries to extend their vocabularies. Moses phrase-based statistical machine translation (Moses PBSMT) er kerfi til þess að þróa og keyra tölfræðilegar vélþýðingar. Hér er dreift fjórum pökkum: 1. Kóða af github geymslusvæði fyrir þjálfun og keyrslu á líkönum 2. Forþjálfuðu is-en vélþýðingarlíkani (Docker) 3. Forþjálfuðu en-is vélþýðingarlíkani (Docker) 4. Framenda til að for- og eftirvinna texta fyrir þýðingar (Docker) Líkönin sem eru sett hér eru ekki (nákvæmlega) þau sömu og voru notuð við mannlegt mat. Þessi líkön hafa aukalega verið þjálfuð á gögnum úr opnum orðabókum til þess að auka orðaforða.
dc.language.iso isl
dc.language.iso eng
dc.publisher Reykjavik University
dc.rights The MIT License (MIT)
dc.rights.uri https://opensource.org/licenses/mit-license.php
dc.rights.label PUB
dc.source.uri https://github.com/cadia-lvl/SMT
dc.subject machine translation
dc.subject statistical machine translation
dc.subject moses
dc.title MT: Moses-SMT (1.0)
dc.type toolService
metashare.ResourceInfo#ContentInfo.detailedType service
metashare.ResourceInfo#ResourceComponentType#ToolServiceInfo.languageDependent true
has.files yes
branding Clarin IS Repository
demo.uri https://nlp.cs.ru.is/moses/translateText
contact.person Haukur Páll Jónsson haukurpj@ru.is Reykjavik University
sponsor Ministry of Education, Science and Culture (Mennta- og menningamálaráðuneytið) Language Technology for Icelandic 2019-2023 Machine Translation - baseline (V3) nationalFunds
files.size 11789798918
files.count 4


 Files in this item

 Download all files in item (10.98 GB)
This item is
Publicly Available
and licensed under:
The MIT License (MIT)
Icon
Name
SMT-master.zip
Size
25.3 MB
Format
application/zip
Description
GitHub repository snapshot
MD5
9f33d407dd2f9d6174587513b07e0fed
 Download file  Preview
 File Preview  
  • SMT-master
    • README.md4 kB
    • .gitignore149 B
    • preprocessing
      • README.md1 kB
      • tests
        • test_api.py184 B
        • list_merging.py1 kB
        • serialization.py1 kB
        • test_read_rmh.py526 B
        • memory_footprint.py304 B
        • test_pipeline.py1 kB
      • docker-build.sh164 B
      • preprocessing
        • client.py750 B
        • file_handler.py4 kB
        • types.py470 B
        • api.py5 kB
        • __init__.py0 B
        • server.py3 kB
        • pipeline.py13 kB
        • resources
          • truecase-model.en13 MB
          • truecase-model.is34 MB
          • __init__.py0 B
          • tok.is7 MB
      • conftest.py0 B
      • requirements.txt1 kB
      • Dockerfile283 B
      • main.py9 kB
      • LICENSE1 kB
    • scripts
      • README.md842 B
      • run_in_singularity.sh764 B
      • 2preprocess
        • preprocess.sh2 kB
        • lm.sh871 B
      • 1format
        • extract_dicts.sh1 kB
        • en_mono_format.py4 kB
      • environment.sh1 kB
      • 4package
        • docker-build.sh543 B
        • README.md1 kB
        • docker-run.sh117 B
        • Dockerfile55 B
      • 3train
        • dict.sh4 kB
        • evaluate.sh514 B
        • translate.sh507 B
      • end_to_end.sh5 kB
      • experiments
        • unkown_tokens.sh1 kB
    • data
      • readme.md1 kB
      • raw
        • parice32 B
        • en_mono39 B
        • rmh37 B
        • dictionary
          • wiki.tsv1 MB
          • manual.tsv27 B
          • apertium-isl-eng.isl-eng.dix2 MB
      • out23 B
      • formatted30 B
    • moses
      • docker-build.sh106 B
      • README.md2 kB
      • Dockerfile1 kB
    • docker-compose.yml617 B
    • LICENSE1 kB
    • notebooks
      • ParIce - 2. filter.ipynb174 kB
      • Moses xmlrpc.ipynb1 kB
      • README.md880 B
      • Moses hand-calculation.ipynb1 kB
      • google_translate.ipynb9 kB
      • data_exploration.ipynb383 kB
      • explore-results.ipynb170 kB
      • ParIce - 1. format.ipynb42 kB
Icon
Name
moses-lvl.tar.gz
Size
493.84 MB
Format
application/gzip
Description
Frontend for translations
MD5
4dfcbb6b3b46a84768a6ca47799e05ae
 Download file  Preview
Icon
Name
moses-smt_is-en.tar.gz
Size
5.3 GB
Format
application/gzip
Description
Trained IS-EN model
MD5
d2dcf5089d7d7cd68a3db46da5eff19c
 Download file  Preview
 File Preview  
Icon
Name
moses-smt_en-is.tar.gz
Size
5.17 GB
Format
application/gzip
Description
Trained EN-IS model
MD5
12099d739edcc710afb7a879ef5a37fc
 Download file  Preview
 File Preview  

Sýna einfalda færslu atriðis