bert-finetuned-ner
This model is a fine-tuned version of BAAI/bge-small-en-v1.5 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0907
- Precision: 0.9049
- Recall: 0.9285
- F1: 0.9165
- Accuracy: 0.9824
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
- distributed_type: multi-GPU
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.0441 | 1.0 | 1250 | 0.0869 | 0.8738 | 0.9148 | 0.8939 | 0.9791 |
| 0.0447 | 2.0 | 2500 | 0.0856 | 0.8851 | 0.9192 | 0.9018 | 0.9802 |
| 0.0477 | 3.0 | 3750 | 0.0852 | 0.8918 | 0.9169 | 0.9042 | 0.9803 |
| 0.0313 | 4.0 | 5000 | 0.0826 | 0.9041 | 0.9283 | 0.9161 | 0.9822 |
| 0.0155 | 5.0 | 6250 | 0.0903 | 0.8987 | 0.9256 | 0.9120 | 0.9812 |
| 0.0147 | 6.0 | 7500 | 0.0864 | 0.9073 | 0.9286 | 0.9178 | 0.9825 |
| 0.0174 | 7.0 | 8750 | 0.0861 | 0.9050 | 0.9265 | 0.9156 | 0.9827 |
| 0.0099 | 8.0 | 10000 | 0.0907 | 0.9049 | 0.9285 | 0.9165 | 0.9824 |
Framework versions
- Transformers 4.51.3
- Pytorch 2.5.1+cu118
- Datasets 3.6.0
- Tokenizers 0.21.4