bert-mini-emotion_classifier
This model is a fine-tuned version of prajjwal1/bert-mini on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0648
- F1: 0.9315
- Roc Auc: 0.9589
- Accuracy: 0.9224
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: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 5
- num_epochs: 1
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 | Roc Auc | Accuracy |
|---|---|---|---|---|---|---|
| 0.4176 | 0.1 | 500 | 0.2929 | 0.6755 | 0.7687 | 0.5550 |
| 0.2278 | 0.19 | 1000 | 0.1623 | 0.8931 | 0.9246 | 0.8630 |
| 0.1513 | 0.29 | 1500 | 0.1184 | 0.9185 | 0.9450 | 0.9022 |
| 0.1198 | 0.38 | 2000 | 0.0957 | 0.9274 | 0.9536 | 0.9197 |
| 0.1011 | 0.48 | 2500 | 0.0815 | 0.9306 | 0.9568 | 0.9230 |
| 0.0881 | 0.58 | 3000 | 0.0729 | 0.9320 | 0.9575 | 0.9237 |
| 0.0815 | 0.67 | 3500 | 0.0669 | 0.9337 | 0.9596 | 0.9256 |
| 0.0767 | 0.77 | 4000 | 0.0633 | 0.9346 | 0.9609 | 0.9260 |
| 0.0721 | 0.86 | 4500 | 0.0612 | 0.9333 | 0.9602 | 0.9233 |
| 0.071 | 0.96 | 5000 | 0.0601 | 0.9339 | 0.9607 | 0.9251 |
Framework versions
- Transformers 4.36.2
- Pytorch 2.1.2+cu121
- Datasets 2.16.0
- Tokenizers 0.15.0