<!-- 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
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: 0.2116
- Bleu: 70.8048
- Gen Len: 77.4185
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 |
|---|---|---|---|---|---|
| No log | 1.0 | 124 | 1.2965 | 18.4921 | 92.7131 |
| No log | 2.0 | 248 | 0.9205 | 33.0814 | 86.4256 |
| No log | 3.0 | 372 | 0.7444 | 40.4913 | 82.542 |
| No log | 4.0 | 496 | 0.6318 | 44.9974 | 82.0258 |
| 1.3399 | 5.0 | 620 | 0.5557 | 49.1838 | 80.1716 |
| 1.3399 | 6.0 | 744 | 0.4952 | 52.3843 | 79.3947 |
| 1.3399 | 7.0 | 868 | 0.4495 | 54.9575 | 78.789 |
| 1.3399 | 8.0 | 992 | 0.4087 | 57.0903 | 78.6098 |
| 0.6118 | 9.0 | 1116 | 0.3782 | 58.8022 | 77.9793 |
| 0.6118 | 10.0 | 1240 | 0.3515 | 60.3209 | 78.5709 |
| 0.6118 | 11.0 | 1364 | 0.3298 | 62.0402 | 77.9904 |
| 0.6118 | 12.0 | 1488 | 0.3097 | 63.6902 | 77.3274 |
| 0.4448 | 13.0 | 1612 | 0.2929 | 64.6001 | 77.624 |
| 0.4448 | 14.0 | 1736 | 0.2782 | 65.6159 | 77.7996 |
| 0.4448 | 15.0 | 1860 | 0.2657 | 66.489 | 78.0329 |
| 0.4448 | 16.0 | 1984 | 0.2563 | 66.9987 | 77.8467 |
| 0.3599 | 17.0 | 2108 | 0.2454 | 67.9398 | 77.6943 |
| 0.3599 | 18.0 | 2232 | 0.2365 | 68.8362 | 77.4833 |
| 0.3599 | 19.0 | 2356 | 0.2303 | 68.9607 | 77.6022 |
| 0.3599 | 20.0 | 2480 | 0.2235 | 69.7223 | 77.4504 |
| 0.3125 | 21.0 | 2604 | 0.2192 | 70.1575 | 77.3618 |
| 0.3125 | 22.0 | 2728 | 0.2161 | 70.3755 | 77.4499 |
| 0.3125 | 23.0 | 2852 | 0.2142 | 70.4898 | 77.4489 |
| 0.3125 | 24.0 | 2976 | 0.2123 | 70.6856 | 77.3623 |
| 0.2875 | 25.0 | 3100 | 0.2116 | 70.8048 | 77.4185 |
Framework versions
- Transformers 4.34.0
- Pytorch 2.1.0+cu118
- Datasets 2.14.5
- Tokenizers 0.14.1
