student-finetuned-REDv2
This model is a fine-tuned version of racai/distilbert-base-romanian-cased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2894
- F1: 0.5107
- Roc Auc: 0.6972
- Accuracy: 0.3996
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: 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: 8
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 | Roc Auc | Accuracy |
|---|---|---|---|---|---|---|
| No log | 1.0 | 256 | 0.3880 | 0.0317 | 0.5090 | 0.0166 |
| 0.4119 | 2.0 | 512 | 0.3440 | 0.2117 | 0.5686 | 0.1381 |
| 0.4119 | 3.0 | 768 | 0.3183 | 0.3701 | 0.6359 | 0.2836 |
| 0.313 | 4.0 | 1024 | 0.3041 | 0.4360 | 0.6653 | 0.3481 |
| 0.313 | 5.0 | 1280 | 0.2974 | 0.4720 | 0.6791 | 0.3702 |
| 0.2758 | 6.0 | 1536 | 0.2926 | 0.4947 | 0.6906 | 0.3886 |
| 0.2758 | 7.0 | 1792 | 0.2908 | 0.4983 | 0.6917 | 0.3904 |
| 0.2571 | 8.0 | 2048 | 0.2894 | 0.5107 | 0.6972 | 0.3996 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.13.1+cu116
- Datasets 2.10.0
- Tokenizers 0.13.2