<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
tuned_cair_five_classes
This model is a fine-tuned version of bert-base-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0976
- F1: 0.9767
- Roc Auc: 0.9815
- Accuracy: 0.9325
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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 20
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 | Roc Auc | Accuracy |
|---|---|---|---|---|---|---|
| No log | 1.0 | 250 | 0.1394 | 0.9480 | 0.9577 | 0.845 |
| 0.1037 | 2.0 | 500 | 0.1221 | 0.95 | 0.9577 | 0.87 |
| 0.1037 | 3.0 | 750 | 0.1107 | 0.9598 | 0.9680 | 0.8875 |
| 0.0593 | 4.0 | 1000 | 0.0965 | 0.9662 | 0.9728 | 0.905 |
| 0.0593 | 5.0 | 1250 | 0.0872 | 0.9734 | 0.9787 | 0.92 |
| 0.0352 | 6.0 | 1500 | 0.0824 | 0.9753 | 0.9802 | 0.925 |
| 0.0352 | 7.0 | 1750 | 0.0906 | 0.9701 | 0.9759 | 0.915 |
| 0.02 | 8.0 | 2000 | 0.0900 | 0.9734 | 0.9787 | 0.925 |
| 0.02 | 9.0 | 2250 | 0.0930 | 0.9727 | 0.9776 | 0.9225 |
| 0.0141 | 10.0 | 2500 | 0.0932 | 0.9734 | 0.9787 | 0.9175 |
| 0.0141 | 11.0 | 2750 | 0.0925 | 0.9760 | 0.9808 | 0.9275 |
| 0.0098 | 12.0 | 3000 | 0.0964 | 0.9741 | 0.9798 | 0.93 |
| 0.0098 | 13.0 | 3250 | 0.0964 | 0.9747 | 0.9798 | 0.9275 |
| 0.0069 | 14.0 | 3500 | 0.0981 | 0.9734 | 0.9787 | 0.925 |
| 0.0069 | 15.0 | 3750 | 0.0930 | 0.9767 | 0.9815 | 0.9325 |
| 0.0058 | 16.0 | 4000 | 0.0939 | 0.9767 | 0.9815 | 0.9325 |
| 0.0058 | 17.0 | 4250 | 0.0959 | 0.9767 | 0.9815 | 0.935 |
| 0.0048 | 18.0 | 4500 | 0.0972 | 0.9753 | 0.9799 | 0.925 |
| 0.0048 | 19.0 | 4750 | 0.0971 | 0.9767 | 0.9815 | 0.9325 |
| 0.0042 | 20.0 | 5000 | 0.0976 | 0.9767 | 0.9815 | 0.9325 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.13.1+cu116
- Datasets 2.8.0
- Tokenizers 0.13.2
