t5-small-finetuned-welsh-plus-br-kw
This model is a fine-tuned version of Jendersen/t5-small-finetuned-welsh-plus-br-kw on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.5220
- Rouge1: 28.9956
- Rouge2: 18.1775
- Rougel: 28.2245
- Rougelsum: 28.2552
- Bleu: 0.017
- Chrf++: 17.2063
- F1: 45.8438
- Gen Len: 17.8503
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: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 4
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Bleu | Chrf++ | F1 | Gen Len |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.4622 | 1.0 | 7853 | 0.5459 | 28.9477 | 18.1192 | 28.1737 | 28.2041 | 0.017 | 17.2025 | 45.8216 | 17.8495 |
| 0.4816 | 2.0 | 15706 | 0.5346 | 28.9406 | 18.1261 | 28.173 | 28.2046 | 0.017 | 17.1972 | 45.8149 | 17.8491 |
| 0.5298 | 3.0 | 23559 | 0.5247 | 28.9843 | 18.1884 | 28.224 | 28.2527 | 0.017 | 17.2122 | 45.8417 | 17.8484 |
| 0.5319 | 4.0 | 31412 | 0.5220 | 28.9956 | 18.1775 | 28.2245 | 28.2552 | 0.017 | 17.2063 | 45.8438 | 17.8503 |
Framework versions
- Transformers 5.13.1
- Pytorch 2.11.0+cu128
- Datasets 4.0.0
- Tokenizers 0.22.2