opus-mt-es-en-GEC-spanish-synthetic-exp
This model is a fine-tuned version of Helsinki-NLP/opus-mt-es-en on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0816
- Gleu: 0.8625
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: 0.0013223040248761528
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 0.09857017609527079
- num_epochs: 4
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Gleu |
|---|---|---|---|---|
| 2.0118 | 1.0 | 2023 | 0.4080 | 0.5965 |
| 0.9899 | 2.0 | 4046 | 0.2020 | 0.7404 |
| 0.5182 | 3.0 | 6069 | 0.1169 | 0.8199 |
| 0.2390 | 4.0 | 8092 | 0.0816 | 0.8625 |
Framework versions
- Transformers 5.5.1
- Pytorch 2.11.0+cu128
- Datasets 4.8.5
- Tokenizers 0.22.2