Edit model card

opus-mt-en-id-open-subtitles

This model was trained from scratch on the open_subtitles dataset. It achieves the following results on the evaluation set:

  • Loss: 2.3148
  • Bleu: 30.2272

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 0.0001
  • train_batch_size: 32
  • eval_batch_size: 32
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 4000
  • num_epochs: 25

Training results

Training Loss Epoch Step Validation Loss Bleu
1.5356 1.0 28125 1.5619 31.8599
1.4703 2.0 56250 1.6047 31.8339
1.3857 3.0 84375 1.6281 32.0796
1.313 4.0 112500 1.6619 31.7391
1.2468 5.0 140625 1.6706 31.9009
1.1831 6.0 168750 1.6924 31.4491
1.1232 7.0 196875 1.7252 31.7229
1.0649 8.0 225000 1.7483 31.7093
1.0078 9.0 253125 1.7697 31.4902
0.9516 10.0 281250 1.8026 31.4342
0.8969 11.0 309375 1.8364 31.2466
0.8436 12.0 337500 1.8747 31.1737
0.7916 13.0 365625 1.9035 31.0118
0.7406 14.0 393750 1.9414 30.9409
0.6912 15.0 421875 1.9776 30.9562
0.6439 16.0 450000 2.0221 30.582
0.5983 17.0 478125 2.0588 30.4478
0.5544 18.0 506250 2.1023 30.4601
0.5126 19.0 534375 2.1367 30.4802
0.474 20.0 562500 2.1790 30.4211
0.438 21.0 590625 2.2131 30.3327
0.4039 22.0 618750 2.2484 30.196
0.3737 23.0 646875 2.2779 30.1145
0.3475 24.0 675000 2.3022 30.2635
0.326 25.0 703125 2.3148 30.2272

Framework versions

  • Transformers 4.26.1
  • Pytorch 2.0.0
  • Datasets 2.10.1
  • Tokenizers 0.11.0
Downloads last month
2

Dataset used to train yonathanstwn/opus-mt-en-id-open-subtitles

Evaluation results