FYP_sq2_mrqa_adqa
This model is a fine-tuned version of kxx-kkk/FYP_sq2_mrqa on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 2.3608
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: 1e-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: 3
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 3.5061 | 0.27 | 500 | 2.6842 |
| 3.3877 | 0.53 | 1000 | 2.5561 |
| 3.3468 | 0.8 | 1500 | 2.5469 |
| 3.3258 | 1.06 | 2000 | 2.4571 |
| 3.2486 | 1.33 | 2500 | 2.4096 |
| 3.2401 | 1.6 | 3000 | 2.3633 |
| 3.2378 | 1.86 | 3500 | 2.3697 |
| 3.2005 | 2.13 | 4000 | 2.3978 |
| 3.1763 | 2.4 | 4500 | 2.3770 |
| 3.1787 | 2.66 | 5000 | 2.3678 |
| 3.164 | 2.93 | 5500 | 2.3608 |
Framework versions
- Transformers 4.39.3
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2