HarmCare_binary
This model is a fine-tuned version of microsoft/deberta-v3-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.9625
- Accuracy: 0.5897
- Precision: 0.5639
- Recall: 0.7916
- F1: 0.6586
- Auc: 0.6309
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: 32
- eval_batch_size: 32
- seed: 1234
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.06
- num_epochs: 8
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 | Auc |
|---|---|---|---|---|---|---|---|---|
| No log | 1.0 | 111 | 0.6878 | 0.5910 | 0.6 | 0.5462 | 0.5718 | 0.6323 |
| No log | 2.0 | 222 | 0.6814 | 0.5541 | 0.5316 | 0.9103 | 0.6712 | 0.6483 |
| No log | 3.0 | 333 | 0.7037 | 0.6003 | 0.5671 | 0.8470 | 0.6794 | 0.6549 |
| No log | 4.0 | 444 | 0.8182 | 0.5897 | 0.5616 | 0.8179 | 0.6660 | 0.6342 |
| 0.5579 | 5.0 | 555 | 0.9625 | 0.5897 | 0.5639 | 0.7916 | 0.6586 | 0.6309 |
Framework versions
- Transformers 4.44.1
- Pytorch 1.11.0
- Datasets 2.12.0
- Tokenizers 0.19.1