<!-- 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. -->
mbart-large-cc25-finetuned-ko-to-en
This model is a fine-tuned version of facebook/mbart-large-cc25 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.9388
- Bleu: 20.301
- Gen Len: 114.7908
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: 2
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 32
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|---|---|---|---|---|---|
| 2.253 | 1.0 | 664 | 1.1693 | 20.073 | 5.8056 |
| 1.1747 | 2.0 | 1328 | 0.9898 | 25.8761 | 7.1737 |
| 0.8827 | 3.0 | 1992 | 0.9286 | 25.4729 | 12.5726 |
| 0.5698 | 4.0 | 2656 | 0.9299 | 18.5817 | 33.1697 |
| 0.4985 | 5.0 | 3320 | 0.9388 | 20.301 | 114.7908 |
Framework versions
- Transformers 4.18.0
- Pytorch 1.11.0+cu113
- Datasets 2.1.0
- Tokenizers 0.12.1
