translator-en-ta
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
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: 0.0003
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.98) and epsilon=1e-09 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 6384
- num_epochs: 15
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|
| 5.8605 | 1.0 | 3548 | 6.1088 |
| 3.8967 | 2.0 | 7096 | 5.2947 |
| 3.1615 | 3.0 | 10644 | 5.0722 |
| 2.8629 | 4.0 | 14192 | 5.0224 |
| 2.6857 | 5.0 | 17740 | 4.9910 |
| 2.5140 | 6.0 | 21288 | 4.9704 |
| 2.4037 | 7.0 | 24836 | 4.9397 |
| 2.2816 | 8.0 | 28384 | 4.9414 |
| 2.1744 | 9.0 | 31932 | 4.9199 |
| 2.0307 | 10.0 | 35480 | 4.8943 |
| 1.9495 | 11.0 | 39028 | 4.8843 |
| 1.8742 | 12.0 | 42576 | 4.8904 |
| 1.8205 | 13.0 | 46124 | 4.8933 |
| 1.7739 | 14.0 | 49672 | 4.9103 |
| 1.7539 | 15.0 | 53220 | 4.8985 |
Framework versions
- Transformers 5.0.0
- Pytorch 2.10.0+cu128
- Datasets 5.0.0
- Tokenizers 0.22.2
