roberta-peft-rating
This model is a fine-tuned version of FacebookAI/roberta-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.9893
- Accuracy: 0.5555
- Mse: 0.6434
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: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Mse |
|---|---|---|---|---|---|
| No log | 1.0 | 425 | 1.3347 | 0.3631 | 1.1621 |
| 1.3948 | 2.0 | 850 | 1.1884 | 0.4617 | 1.0093 |
| 1.2586 | 3.0 | 1275 | 1.0602 | 0.5153 | 0.6819 |
| 1.0752 | 4.0 | 1700 | 1.0236 | 0.5301 | 0.7255 |
| 1.028 | 5.0 | 2125 | 1.0017 | 0.5428 | 0.6455 |
| 1.0183 | 6.0 | 2550 | 0.9974 | 0.5527 | 0.6541 |
| 1.0183 | 7.0 | 2975 | 0.9956 | 0.5514 | 0.6307 |
| 1.0076 | 8.0 | 3400 | 0.9925 | 0.5527 | 0.6510 |
| 0.9974 | 9.0 | 3825 | 0.9914 | 0.5589 | 0.6427 |
| 0.9863 | 10.0 | 4250 | 0.9893 | 0.5555 | 0.6434 |
Framework versions
- PEFT 0.13.2
- Transformers 4.44.2
- Pytorch 2.5.0+cu121
- Datasets 3.1.0
- Tokenizers 0.19.1