metadata
language: inc
tags:
- translation
license: apache-2.0
inc-inc
source languages: inc
target languages: inc
OPUS readme: inc-inc
dataset: opus
model: transformer
source language(s): asm hin mar urd
target language(s): asm hin mar urd
model: transformer
pre-processing: normalization + SentencePiece (spm32k,spm32k)
a sentence initial language token is required in the form of
>>id<<
(id = valid target language ID)download original weights: opus-2020-07-27.zip
test set translations: opus-2020-07-27.test.txt
test set scores: opus-2020-07-27.eval.txt
Benchmarks
testset | BLEU | chr-F |
---|---|---|
Tatoeba-test.asm-hin.asm.hin | 2.6 | 0.231 |
Tatoeba-test.hin-asm.hin.asm | 9.1 | 0.262 |
Tatoeba-test.hin-mar.hin.mar | 28.1 | 0.548 |
Tatoeba-test.hin-urd.hin.urd | 19.9 | 0.508 |
Tatoeba-test.mar-hin.mar.hin | 11.6 | 0.466 |
Tatoeba-test.multi.multi | 17.1 | 0.464 |
Tatoeba-test.urd-hin.urd.hin | 13.5 | 0.377 |