bge-small-en-v1.5-conll2003-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: 0.1056
- Precision: 0.9121
- Recall: 0.9322
- F1: 0.9220
- Accuracy: 0.9829
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: 5e-05
- train_batch_size: 16
- eval_batch_size: 64
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 20
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.1888 | 1.0 | 625 | 0.1287 | 0.8204 | 0.8777 | 0.8480 | 0.9724 |
| 0.0831 | 2.0 | 1250 | 0.0827 | 0.8884 | 0.9133 | 0.9007 | 0.9803 |
| 0.0740 | 3.0 | 1875 | 0.0815 | 0.8881 | 0.9125 | 0.9001 | 0.9795 |
| 0.0304 | 4.0 | 2500 | 0.0892 | 0.9019 | 0.9221 | 0.9119 | 0.9810 |
| 0.0281 | 5.0 | 3125 | 0.0805 | 0.9040 | 0.9273 | 0.9155 | 0.9819 |
| 0.0228 | 6.0 | 3750 | 0.0862 | 0.9086 | 0.9280 | 0.9182 | 0.9822 |
| 0.0149 | 7.0 | 4375 | 0.0859 | 0.9059 | 0.9303 | 0.9180 | 0.9819 |
| 0.0163 | 8.0 | 5000 | 0.0864 | 0.9127 | 0.9286 | 0.9206 | 0.9831 |
| 0.0096 | 9.0 | 5625 | 0.0974 | 0.9037 | 0.9273 | 0.9154 | 0.9815 |
| 0.0091 | 10.0 | 6250 | 0.1028 | 0.9148 | 0.9285 | 0.9216 | 0.9819 |
| 0.0105 | 11.0 | 6875 | 0.0981 | 0.9090 | 0.9325 | 0.9206 | 0.9822 |
| 0.0055 | 12.0 | 7500 | 0.1038 | 0.9129 | 0.9275 | 0.9201 | 0.9824 |
| 0.0040 | 13.0 | 8125 | 0.1066 | 0.9174 | 0.9340 | 0.9256 | 0.9831 |
| 0.0059 | 14.0 | 8750 | 0.1059 | 0.9174 | 0.9325 | 0.9249 | 0.9830 |
| 0.0012 | 15.0 | 9375 | 0.1090 | 0.9182 | 0.9300 | 0.9241 | 0.9829 |
| 0.0028 | 16.0 | 10000 | 0.1056 | 0.9121 | 0.9322 | 0.9220 | 0.9829 |
Framework versions
- Transformers 5.18.0
- Pytorch 2.11.0+cu130
- Datasets 4.8.5
- Tokenizers 0.23.2