bert_large_mimic_iii_token_classification_top15_def
This model is a fine-tuned version of bert-large-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.8273
- Precision Macro: 0.1871
- Recall Macro: 0.1912
- F1 Macro: 0.1847
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: 2
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 4
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.3
- num_epochs: 8
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision Macro | Recall Macro | F1 Macro |
|---|---|---|---|---|---|---|
| 0.8795 | 1.0 | 2129 | 0.7131 | 0.5623 | 0.1055 | 0.0931 |
| 0.7993 | 2.0 | 4258 | 0.6742 | 0.2063 | 0.1338 | 0.1231 |
| 0.7907 | 3.0 | 6387 | 0.6615 | 0.2553 | 0.1081 | 0.1031 |
| 0.7417 | 4.0 | 8516 | 0.6729 | 0.2978 | 0.1404 | 0.1417 |
| 0.685 | 5.0 | 10645 | 0.6860 | 0.1878 | 0.1479 | 0.1471 |
| 0.6222 | 6.0 | 12774 | 0.7049 | 0.1945 | 0.1757 | 0.1702 |
| 0.5368 | 7.0 | 14903 | 0.7634 | 0.1875 | 0.1784 | 0.1754 |
| 0.5082 | 8.0 | 17032 | 0.8273 | 0.1871 | 0.1912 | 0.1847 |
Framework versions
- Transformers 4.34.0.dev0
- Pytorch 2.0.1+cu118
- Datasets 2.14.6.dev0
- Tokenizers 0.14.0