QA_RoBERTa
This model was trained from scratch on the None dataset. It achieves the following results on the evaluation set:
- Loss: 2.7080
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: 320
- eval_batch_size: 5
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 40
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| No log | 1.0 | 1 | 6.2494 |
| No log | 2.0 | 2 | 6.1948 |
| No log | 3.0 | 3 | 6.1225 |
| No log | 4.0 | 4 | 6.0169 |
| No log | 5.0 | 5 | 5.8711 |
| No log | 6.0 | 6 | 5.6949 |
| No log | 7.0 | 7 | 5.4383 |
| No log | 8.0 | 8 | 5.1985 |
| No log | 9.0 | 9 | 5.1196 |
| No log | 10.0 | 10 | 5.1005 |
| No log | 11.0 | 11 | 4.9257 |
| No log | 12.0 | 12 | 4.5831 |
| No log | 13.0 | 13 | 4.3068 |
| No log | 14.0 | 14 | 4.1781 |
| No log | 15.0 | 15 | 4.0040 |
| No log | 16.0 | 16 | 3.8976 |
| No log | 17.0 | 17 | 3.8215 |
| No log | 18.0 | 18 | 3.7631 |
| No log | 19.0 | 19 | 3.6786 |
| No log | 20.0 | 20 | 3.5771 |
| No log | 21.0 | 21 | 3.5172 |
| No log | 22.0 | 22 | 3.4808 |
| No log | 23.0 | 23 | 3.4825 |
| No log | 24.0 | 24 | 3.5539 |
| No log | 25.0 | 25 | 3.5337 |
| No log | 26.0 | 26 | 3.4939 |
| No log | 27.0 | 27 | 3.5150 |
| No log | 28.0 | 28 | 3.5459 |
| No log | 29.0 | 29 | 3.5461 |
| No log | 30.0 | 30 | 3.4210 |
| No log | 31.0 | 31 | 3.2489 |
| No log | 32.0 | 32 | 3.1328 |
| No log | 33.0 | 33 | 3.1024 |
| No log | 34.0 | 34 | 3.0777 |
| No log | 35.0 | 35 | 3.0049 |
| No log | 36.0 | 36 | 2.9391 |
| No log | 37.0 | 37 | 2.8754 |
| No log | 38.0 | 38 | 2.8461 |
| No log | 39.0 | 39 | 2.7858 |
| No log | 40.0 | 40 | 2.7080 |
Framework versions
- Transformers 4.36.2
- Pytorch 2.1.2+cu121
- Datasets 2.14.7
- Tokenizers 0.15.0
- Downloads last month
- -