<!-- 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_30
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: 0.6220
- Precision Macro: 0.1301
- Recall Macro: 0.1650
- F1 Macro: 0.1397
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: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.2
- num_epochs: 50
Training results
Training Loss | Epoch | Step | Validation Loss | Precision Macro | Recall Macro | F1 Macro |
---|---|---|---|---|---|---|
0.9646 | 1.0 | 241 | 0.9086 | 0.0 | 0.0 | 0.0 |
0.6421 | 2.0 | 482 | 0.6963 | 0.0425 | 0.0634 | 0.0507 |
0.6868 | 3.0 | 723 | 0.6505 | 0.0444 | 0.0771 | 0.0559 |
0.5139 | 4.0 | 964 | 0.6178 | 0.0941 | 0.0929 | 0.0918 |
0.5583 | 5.0 | 1205 | 0.5899 | 0.1065 | 0.1279 | 0.1103 |
0.6335 | 6.0 | 1446 | 0.5899 | 0.1495 | 0.1298 | 0.1265 |
0.5661 | 7.0 | 1687 | 0.5769 | 0.1016 | 0.1397 | 0.1162 |
0.4091 | 8.0 | 1928 | 0.5865 | 0.1359 | 0.1429 | 0.1260 |
0.5056 | 9.0 | 2169 | 0.5989 | 0.1118 | 0.1186 | 0.1065 |
0.427 | 10.0 | 2410 | 0.6220 | 0.1301 | 0.1650 | 0.1397 |
Framework versions
- Transformers 4.34.0.dev0
- Pytorch 2.0.1+cu118
- Datasets 2.14.6.dev0
- Tokenizers 0.14.0