roberta-base-finetuned-academic
This model is a fine-tuned version of roberta-base on the elsevier-oa-cc-by dataset. It achieves the following results on the evaluation set:
- Loss: 2.1158
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: 0.0001
- 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: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.1903 | 0.25 | 1025 | 2.0998 |
| 2.1752 | 0.5 | 2050 | 2.1186 |
| 2.1864 | 0.75 | 3075 | 2.1073 |
| 2.1874 | 1.0 | 4100 | 2.1177 |
| 2.1669 | 1.25 | 5125 | 2.1091 |
| 2.1859 | 1.5 | 6150 | 2.1212 |
| 2.1783 | 1.75 | 7175 | 2.1096 |
| 2.1734 | 2.0 | 8200 | 2.0998 |
| 2.1712 | 2.25 | 9225 | 2.0972 |
| 2.1812 | 2.5 | 10250 | 2.1051 |
| 2.1811 | 2.75 | 11275 | 2.1150 |
| 2.1826 | 3.0 | 12300 | 2.1097 |
| 2.172 | 3.25 | 13325 | 2.1115 |
| 2.1745 | 3.5 | 14350 | 2.1098 |
| 2.1758 | 3.75 | 15375 | 2.1101 |
| 2.1834 | 4.0 | 16400 | 2.1232 |
| 2.1836 | 4.25 | 17425 | 2.1052 |
| 2.1791 | 4.5 | 18450 | 2.1186 |
| 2.172 | 4.75 | 19475 | 2.1039 |
| 2.1797 | 5.0 | 20500 | 2.1015 |
Framework versions
- Transformers 4.19.0.dev0
- Pytorch 1.11.0+cu113
- Datasets 2.3.2
- Tokenizers 0.12.1