variant_tapt_freeze_llrd_LR_5e
This model is a fine-tuned version of Mardiyyah/variant-tapt_freeze_llrd-LR_5e-05 on the OTAR3088/Variants_IOB-V2 dataset. It achieves the following results on the evaluation set:
- Loss: 0.0651
- Precision: 0.8584
- Recall: 0.8664
- F1: 0.8624
- Accuracy: 0.9858
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
- 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: 0.1
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.5514 | 1.0 | 34 | 0.2347 | 0.2767 | 0.2028 | 0.2340 | 0.9168 |
| 0.1451 | 2.0 | 68 | 0.0934 | 0.6139 | 0.7327 | 0.6681 | 0.9716 |
| 0.0725 | 3.0 | 102 | 0.0718 | 0.8148 | 0.8111 | 0.8129 | 0.9812 |
| 0.0485 | 4.0 | 136 | 0.0651 | 0.8136 | 0.8249 | 0.8192 | 0.9831 |
| 0.0321 | 5.0 | 170 | 0.0605 | 0.8378 | 0.8571 | 0.8474 | 0.9850 |
| 0.0243 | 6.0 | 204 | 0.0646 | 0.8493 | 0.8571 | 0.8532 | 0.9852 |
| 0.0181 | 7.0 | 238 | 0.0641 | 0.8468 | 0.8664 | 0.8565 | 0.9864 |
| 0.0161 | 8.0 | 272 | 0.0603 | 0.8584 | 0.8664 | 0.8624 | 0.9858 |
| 0.0125 | 9.0 | 306 | 0.0584 | 0.8386 | 0.8618 | 0.85 | 0.9858 |
| 0.0118 | 10.0 | 340 | 0.0616 | 0.8462 | 0.8618 | 0.8539 | 0.9862 |
Framework versions
- Transformers 4.48.2
- Pytorch 2.4.1+cu121
- Datasets 3.0.2
- Tokenizers 0.21.0
- Downloads last month
- 6