bert-crossencoder-focal_loss
This model is a fine-tuned version of google-bert/bert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.1413
- Accuracy: 0.5858
- Precision: 0.5911
- Recall: 0.5858
- F1: 0.5848
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: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- num_epochs: 7
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| 0.1669 | 1.0 | 78 | 0.1589 | 0.4628 | 0.3807 | 0.4628 | 0.3868 |
| 0.1378 | 2.0 | 156 | 0.1398 | 0.5243 | 0.5555 | 0.5243 | 0.4713 |
| 0.1111 | 3.0 | 234 | 0.1255 | 0.5955 | 0.6017 | 0.5955 | 0.5947 |
| 0.0854 | 4.0 | 312 | 0.1244 | 0.5858 | 0.5909 | 0.5858 | 0.5840 |
| 0.0484 | 5.0 | 390 | 0.1316 | 0.5761 | 0.5871 | 0.5761 | 0.5722 |
| 0.046 | 6.0 | 468 | 0.1375 | 0.5987 | 0.6002 | 0.5987 | 0.5992 |
| 0.0325 | 7.0 | 546 | 0.1413 | 0.5858 | 0.5911 | 0.5858 | 0.5848 |
Framework versions
- Transformers 4.45.1
- Pytorch 2.4.0
- Datasets 3.0.1
- Tokenizers 0.20.0