<!-- 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. -->
opus-mt-finetuned-jv-to-id
This model is a fine-tuned version of hermanshid/opus-mt-finetuned-id-to-jv on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.6790
- Bleu: 45.3883
- Gen Len: 57.2188
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: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|---|---|---|---|---|---|
| 0.8743 | 1.0 | 2500 | 0.8210 | 40.2457 | 57.0176 |
| 0.7743 | 2.0 | 5000 | 0.7618 | 42.225 | 56.9092 |
| 0.7225 | 3.0 | 7500 | 0.7329 | 43.3571 | 57.0074 |
| 0.6783 | 4.0 | 10000 | 0.7116 | 44.2165 | 57.084 |
| 0.6467 | 5.0 | 12500 | 0.6995 | 44.5317 | 57.0166 |
| 0.6268 | 6.0 | 15000 | 0.6919 | 44.8921 | 57.1716 |
| 0.61 | 7.0 | 17500 | 0.6854 | 45.162 | 57.1296 |
| 0.5956 | 8.0 | 20000 | 0.6819 | 45.2834 | 57.2868 |
| 0.5816 | 9.0 | 22500 | 0.6789 | 45.3757 | 57.2016 |
| 0.5717 | 10.0 | 25000 | 0.6790 | 45.3883 | 57.2188 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3
