OPUS-MT-train/models/en-tvl
2020-05-29 14:43:36 +03:00
..
README.md fixed a bug in eval-testsets 2020-05-29 14:43:36 +03:00

Tatoeba-2020-01-17.zip

Benchmarks

testset BLEU chr-F
JW300.en.tvl 46.9 0.625

opus-2020-01-20.zip

Benchmarks

testset BLEU chr-F
JW300.en.tvl 46.9 0.625

opus+bt-2020-05-23.zip

Training data: opus+bt

  • en-tvl: Tatoeba (14)
  • en-tvl: total size = 14
  • unused dev/test data is added to training data
  • total size (opus+bt): 175142

Validation data

  • en-tvl: JW300

  • devset = top 2500 lines of JW300.src.shuffled!

  • testset = next 2500 lines of JW300.src.shuffled!

  • remaining lines are added to traindata

Benchmarks

testset BLEU chr-F
JW300.en.tvl 47.7 0.630