roberta-base-finetuned-squad
This model is a fine-tuned version of roberta-base on the squad dataset. It achieves the following results on the evaluation set:
- Loss: 1.1119
- Exact match: 85.04257
- F1 score: 91.82363
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: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.8806 | 1.0 | 5536 | 0.8635 |
| 0.7139 | 2.0 | 11072 | 0.8431 |
| 0.5514 | 3.0 | 16608 | 0.9000 |
| 0.4146 | 4.0 | 22144 | 1.0211 |
| 0.3367 | 5.0 | 27680 | 1.1119 |
Framework versions
- Transformers 4.34.0
- Pytorch 1.12.1
- Datasets 2.14.5
- Tokenizers 0.14.1