<!-- 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_large_ner_model_mimic_top5
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: 0.5861
- Precision Macro: 0.2520
- Recall Macro: 0.2807
- F1 Macro: 0.2650
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: 4
- 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: 8
Training results
Training Loss | Epoch | Step | Validation Loss | Precision Macro | Recall Macro | F1 Macro |
---|---|---|---|---|---|---|
0.556 | 1.0 | 1549 | 0.4738 | 0.1245 | 0.1562 | 0.1384 |
0.4855 | 2.0 | 3098 | 0.4485 | 0.2358 | 0.2378 | 0.2297 |
0.4798 | 3.0 | 4647 | 0.4261 | 0.2055 | 0.2640 | 0.2182 |
0.4445 | 4.0 | 6196 | 0.4239 | 0.2159 | 0.2426 | 0.2182 |
0.3877 | 5.0 | 7745 | 0.4471 | 0.2696 | 0.2596 | 0.2579 |
0.342 | 6.0 | 9294 | 0.4849 | 0.2401 | 0.2823 | 0.2536 |
0.267 | 7.0 | 10843 | 0.5234 | 0.2577 | 0.2871 | 0.2707 |
0.2439 | 8.0 | 12392 | 0.5861 | 0.2520 | 0.2807 | 0.2650 |
Framework versions
- Transformers 4.34.0.dev0
- Pytorch 2.0.1+cu118
- Datasets 2.14.6.dev0
- Tokenizers 0.14.0