-
source language name: Portuguese
-
target language name: Catalan
-
-
model: transformer
-
source language code: pt
-
target language code: ca
-
dataset: opus
-
release date: 2021-02-23
-
pre-processing: normalization + SentencePiece (spm32k,spm32k)
-
download original weights:
opus-2021-02-23.zip
-
Training data:
- pob-cat: Tatoeba-train (406704)
- por-cat: Tatoeba-train (756629)
-
Validation data:
- cat-pob: Tatoeba-dev, 356
- cat-por: Tatoeba-dev, 644
- total-size-shuffled: 999
- devset-selected: top 999 lines of Tatoeba-dev.src.shuffled!
-
Test data:
- Tatoeba-test.por-cat: 729/5878
-
test set translations file:
test.txt
-
test set scores file:
eval.txt
-
BLEU-scores
|Test set|score|
|---|---|
|Tatoeba-test.por-cat|41.5|
-
chr-F-scores
|Test set|score|
|---|---|
|Tatoeba-test.por-cat|0.633|