<!-- 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. -->
ko_en_translation_9
This model is a fine-tuned version of inhee/opus-mt-ko-en-finetuned-ko-to-en5 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.8852
- Bleu: 45.395
- Gen Len: 38.8647
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: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 7
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|---|---|---|---|---|---|
| 1.2529 | 1.0 | 4900 | 1.0958 | 40.5188 | 38.6684 |
| 1.0565 | 2.0 | 9800 | 0.9885 | 42.7968 | 38.6141 |
| 0.9518 | 3.0 | 14700 | 0.9370 | 43.8495 | 38.7762 |
| 0.8798 | 4.0 | 19600 | 0.9119 | 44.7342 | 38.7903 |
| 0.8401 | 5.0 | 24500 | 0.8958 | 45.2518 | 38.8909 |
| 0.8075 | 6.0 | 29400 | 0.8883 | 45.326 | 38.8503 |
| 0.7934 | 7.0 | 34300 | 0.8852 | 45.395 | 38.8647 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.0
- Datasets 2.1.0
- Tokenizers 0.13.3
