<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
fugumt-en-ja-finetuned-en-to-ja-21939-mod512
This model is a fine-tuned version of staka/fugumt-en-ja on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0712
- Bleu: 81.8494
- Gen Len: 17.3121
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: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 25
Training results
Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
---|---|---|---|---|---|
1.4245 | 1.0 | 1372 | 1.0184 | 35.2708 | 17.3468 |
1.0059 | 2.0 | 2744 | 0.7893 | 46.564 | 16.9338 |
0.805 | 3.0 | 4116 | 0.6361 | 52.2901 | 17.094 |
0.6995 | 4.0 | 5488 | 0.5297 | 56.2805 | 17.2441 |
0.5935 | 5.0 | 6860 | 0.4477 | 59.6019 | 17.2393 |
0.5285 | 6.0 | 8232 | 0.3830 | 62.019 | 17.2745 |
0.4741 | 7.0 | 9604 | 0.3277 | 64.7888 | 17.2489 |
0.4259 | 8.0 | 10976 | 0.2836 | 67.0347 | 17.2363 |
0.3816 | 9.0 | 12348 | 0.2467 | 68.5365 | 17.2635 |
0.3555 | 10.0 | 13720 | 0.2183 | 70.4964 | 17.2489 |
0.311 | 11.0 | 15092 | 0.1906 | 72.0133 | 17.3511 |
0.2807 | 12.0 | 16464 | 0.1686 | 73.993 | 17.2167 |
0.2651 | 13.0 | 17836 | 0.1510 | 75.411 | 17.1765 |
0.2513 | 14.0 | 19208 | 0.1346 | 75.6697 | 17.3105 |
0.2398 | 15.0 | 20580 | 0.1220 | 77.2019 | 17.2747 |
0.2158 | 16.0 | 21952 | 0.1109 | 78.1196 | 17.2447 |
0.2068 | 17.0 | 23324 | 0.1010 | 78.8838 | 17.3506 |
0.1948 | 18.0 | 24696 | 0.0939 | 79.4503 | 17.3367 |
0.1881 | 19.0 | 26068 | 0.0878 | 80.1415 | 17.2807 |
0.1738 | 20.0 | 27440 | 0.0825 | 80.6082 | 17.3224 |
0.1705 | 21.0 | 28812 | 0.0784 | 80.8747 | 17.3527 |
0.1596 | 22.0 | 30184 | 0.0756 | 81.5624 | 17.2991 |
0.156 | 23.0 | 31556 | 0.0733 | 81.7063 | 17.313 |
0.1529 | 24.0 | 32928 | 0.0716 | 81.769 | 17.3133 |
0.146 | 25.0 | 34300 | 0.0712 | 81.8494 | 17.3121 |
Framework versions
- Transformers 4.34.0
- Pytorch 2.1.0+cu118
- Datasets 2.14.5
- Tokenizers 0.14.1