cat-spa

opus-2020-06-15.zip

• dataset: opus
• model: transformer-align
• source language(s): cat
• target language(s): spa
• model: transformer-align
• pre-processing: normalization + SentencePiece (spm32k,spm32k)
• download: opus-2020-06-15.zip
• test set translations: opus-2020-06-15.test.txt
• test set scores: opus-2020-06-15.eval.txt

Training data: opus

• cat-spa: Tatoeba-train (7953020)
• cat-spa: total size = 7953020
• total size (opus): 7949148

Validation data

• cat-spa: Tatoeba-dev, 1000

• total size of shuffled dev data: 1000

• devset = top 1000 lines of Tatoeba-dev.src.shuffled!

Benchmarks

testset	BLEU	chr-F
Tatoeba-test.cat.spa	73.9	0.856

opus-2020-06-17.zip

• dataset: opus
• model: transformer-align
• source language(s): cat
• target language(s): spa
• model: transformer-align
• pre-processing: normalization + SentencePiece (spm32k,spm32k)
• download: opus-2020-06-17.zip
• test set translations: opus-2020-06-17.test.txt
• test set scores: opus-2020-06-17.eval.txt

Benchmarks

testset	BLEU	chr-F
Tatoeba-test.cat.spa	73.9	0.856

opus-2020-09-10.zip

• dataset: opus
• model: transformer-align
• source language(s): cat
• target language(s): spa
• model: transformer-align
• pre-processing: normalization + SentencePiece (spm32k,spm32k)
• download: opus-2020-09-10.zip
• test set translations: opus-2020-09-10.test.txt
• test set scores: opus-2020-09-10.eval.txt

Benchmarks

testset	BLEU	chr-F
Tatoeba-test.cat.spa	73.9	0.856