<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
bert_ner_model_mimic_3
This model is a fine-tuned version of bert-base-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.7115
- Precision Macro: 0.1745
- Recall Macro: 0.1333
- F1 Macro: 0.1182
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: 0.0002
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.3
- num_epochs: 50
Training results
Training Loss | Epoch | Step | Validation Loss | Precision Macro | Recall Macro | F1 Macro |
---|---|---|---|---|---|---|
No log | 1.0 | 482 | 0.6468 | 0.0437 | 0.0780 | 0.0556 |
1.1127 | 2.0 | 964 | 0.6040 | 0.0889 | 0.1120 | 0.0962 |
0.611 | 3.0 | 1446 | 0.5957 | 0.1426 | 0.1122 | 0.1105 |
0.5681 | 4.0 | 1928 | 0.5876 | 0.1239 | 0.1171 | 0.1148 |
0.5514 | 5.0 | 2410 | 0.6214 | 0.1399 | 0.1171 | 0.1124 |
0.5186 | 6.0 | 2892 | 0.6087 | 0.1350 | 0.1322 | 0.1195 |
0.4952 | 7.0 | 3374 | 0.6115 | 0.1381 | 0.1420 | 0.1291 |
0.475 | 8.0 | 3856 | 0.6665 | 0.1356 | 0.1223 | 0.1174 |
0.4632 | 9.0 | 4338 | 0.6593 | 0.1405 | 0.1476 | 0.1356 |
0.4378 | 10.0 | 4820 | 0.6430 | 0.1406 | 0.1401 | 0.1312 |
0.4288 | 11.0 | 5302 | 0.6485 | 0.1394 | 0.1353 | 0.1293 |
0.4369 | 12.0 | 5784 | 0.6722 | 0.1198 | 0.1249 | 0.1129 |
0.4215 | 13.0 | 6266 | 0.6769 | 0.1243 | 0.1357 | 0.1235 |
0.4274 | 14.0 | 6748 | 0.7115 | 0.1745 | 0.1333 | 0.1182 |
Framework versions
- Transformers 4.34.0.dev0
- Pytorch 2.0.1+cu118
- Datasets 2.14.6.dev0
- Tokenizers 0.14.0