bert-mini-url
This model is a fine-tuned version of prajjwal1/bert-mini on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0565
- Accuracy: 0.9873
- Precision: 0.9848
- Recall: 0.9912
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: 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: 4
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall |
|---|---|---|---|---|---|---|
| 0.0644 | 1.0 | 32322 | 0.0633 | 0.9815 | 0.9832 | 0.9818 |
| 0.0579 | 2.0 | 64644 | 0.0572 | 0.9853 | 0.9818 | 0.9906 |
| 0.0485 | 3.0 | 96966 | 0.0564 | 0.9867 | 0.9859 | 0.9892 |
| 0.0439 | 4.0 | 129288 | 0.0565 | 0.9873 | 0.9848 | 0.9912 |
Framework versions
- Transformers 4.40.2
- Pytorch 2.3.0+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1