FYP_sq2_mrqa_adqa_synqa
This model is a fine-tuned version of kxx-kkk/FYP_sq2_mrqa_adqa on the syn_qa dataset. It achieves the following results on the evaluation set:
- Loss: 0.9362
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: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 2
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.82 | 0.25 | 2000 | 1.1116 |
| 1.612 | 0.51 | 4000 | 1.0270 |
| 1.505 | 0.76 | 6000 | 0.9874 |
| 1.4476 | 1.01 | 8000 | 0.9784 |
| 1.3898 | 1.27 | 10000 | 0.9476 |
| 1.3544 | 1.52 | 12000 | 0.9368 |
| 1.3278 | 1.78 | 14000 | 0.9362 |
Framework versions
- Transformers 4.39.3
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2