CeLLaTe-ner-2class-bioformer16l-with-gazetteers-lr_2.103e-5
This model is a fine-tuned version of bioformers/bioformer-16l on the OTAR3088/CeLLaTe-ner-2class-iob_final dataset. It achieves the following results on the evaluation set:
- Loss: 0.0796
- Precision: 0.7699
- Recall: 0.7871
- Micro F1: 0.7784
- Weighted F1: 0.7788
- Macro F1: 0.7898
- Accuracy: 0.9820
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: 2.1033235284928615e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 17
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.02
- num_epochs: 20
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | Micro F1 | Weighted F1 | Macro F1 | Accuracy |
|---|---|---|---|---|---|---|---|---|---|
| 0.1577 | 1.0 | 1052 | 0.0624 | 0.7736 | 0.7504 | 0.7619 | 0.7619 | 0.7703 | 0.9816 |
| 0.0202 | 2.0 | 2104 | 0.0732 | 0.7993 | 0.7589 | 0.7786 | 0.7788 | 0.7894 | 0.9820 |
| 0.0099 | 3.0 | 3156 | 0.0794 | 0.7699 | 0.7871 | 0.7784 | 0.7788 | 0.7898 | 0.9820 |
| 0.0059 | 4.0 | 4208 | 0.0947 | 0.7329 | 0.7903 | 0.7605 | 0.7608 | 0.7726 | 0.9802 |
| 0.0038 | 5.0 | 5260 | 0.1122 | 0.8108 | 0.7455 | 0.7768 | 0.7767 | 0.7865 | 0.9824 |
| 0.0027 | 6.0 | 6312 | 0.1080 | 0.7710 | 0.7558 | 0.7633 | 0.7636 | 0.7723 | 0.9816 |
Framework versions
- Transformers 4.48.2
- Pytorch 2.4.1+cu121
- Datasets 3.0.2
- Tokenizers 0.21.0