CeLLaTe-ner-2class-tapt-bioformer16l-tokenizer-original-baseline
This model is a fine-tuned version of Mardiyyah/CeLLaTe-tapt-bioformer16l-tokenizer-original-baseline on the OTAR3088/CeLLaTe-ner-2class-iob_final dataset. It achieves the following results on the evaluation set:
- Loss: 0.0934
- Precision: 0.8061
- Recall: 0.7437
- Micro F1: 0.7737
- Weighted F1: 0.7736
- Macro F1: 0.7811
- 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: 16
- eval_batch_size: 16
- seed: 3407
- 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: 0.01
- 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 |
|---|---|---|---|---|---|---|---|---|---|
| 0.2626 | 1.0 | 263 | 0.0682 | 0.6586 | 0.6825 | 0.6703 | 0.6698 | 0.6649 | 0.9780 |
| 0.048 | 2.0 | 526 | 0.0622 | 0.7681 | 0.7227 | 0.7447 | 0.7451 | 0.7543 | 0.9807 |
| 0.0332 | 3.0 | 789 | 0.0667 | 0.7394 | 0.6968 | 0.7175 | 0.7181 | 0.7262 | 0.9792 |
| 0.0251 | 4.0 | 1052 | 0.0656 | 0.7903 | 0.7232 | 0.7553 | 0.7554 | 0.7624 | 0.9816 |
| 0.0194 | 5.0 | 1315 | 0.0725 | 0.7613 | 0.7330 | 0.7469 | 0.7468 | 0.7461 | 0.9813 |
| 0.0157 | 6.0 | 1578 | 0.0908 | 0.7707 | 0.6449 | 0.7022 | 0.7023 | 0.7049 | 0.9789 |
| 0.0133 | 7.0 | 1841 | 0.0781 | 0.7826 | 0.7406 | 0.7610 | 0.7613 | 0.7690 | 0.9819 |
| 0.0107 | 8.0 | 2104 | 0.0888 | 0.8054 | 0.7165 | 0.7583 | 0.7583 | 0.7632 | 0.9818 |
| 0.0091 | 9.0 | 2367 | 0.0852 | 0.7921 | 0.7446 | 0.7676 | 0.7677 | 0.7727 | 0.9819 |
| 0.0077 | 10.0 | 2630 | 0.0935 | 0.8061 | 0.7437 | 0.7737 | 0.7736 | 0.7811 | 0.9824 |
| 0.0066 | 11.0 | 2893 | 0.0937 | 0.7766 | 0.7370 | 0.7563 | 0.7569 | 0.7669 | 0.9813 |
| 0.0059 | 12.0 | 3156 | 0.1040 | 0.8135 | 0.7294 | 0.7692 | 0.7692 | 0.7778 | 0.9820 |
| 0.0051 | 13.0 | 3419 | 0.1036 | 0.7753 | 0.7285 | 0.7512 | 0.7514 | 0.7571 | 0.9812 |
| 0.0046 | 14.0 | 3682 | 0.1075 | 0.8038 | 0.7294 | 0.7648 | 0.7649 | 0.7724 | 0.9818 |
| 0.0043 | 15.0 | 3945 | 0.1113 | 0.8080 | 0.7303 | 0.7672 | 0.7673 | 0.7767 | 0.9818 |
Framework versions
- Transformers 4.48.2
- Pytorch 2.4.1+cu121
- Datasets 3.0.2
- Tokenizers 0.21.0