BioBERT-LitCovid-v1.3hh
This model is a fine-tuned version of dmis-lab/biobert-v1.1 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.9050
- Hamming loss: 0.0147
- F1 micro: 0.8717
- F1 macro: 0.4368
- F1 weighted: 0.8882
- F1 samples: 0.8857
- Precision micro: 0.8176
- Precision macro: 0.3560
- Precision weighted: 0.8520
- Precision samples: 0.8728
- Recall micro: 0.9334
- Recall macro: 0.7011
- Recall weighted: 0.9334
- Recall samples: 0.9438
- Roc Auc: 0.9608
- Accuracy: 0.7014
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: 5e-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: 0.11492820779210673
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Hamming loss | F1 micro | F1 macro | F1 weighted | F1 samples | Precision micro | Precision macro | Precision weighted | Precision samples | Recall micro | Recall macro | Recall weighted | Recall samples | Roc Auc | Accuracy |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
1.1889 | 1.0 | 2272 | 0.4213 | 0.0512 | 0.6596 | 0.2446 | 0.8084 | 0.7608 | 0.5126 | 0.1941 | 0.7385 | 0.7077 | 0.9250 | 0.8376 | 0.9250 | 0.9404 | 0.9376 | 0.4492 |
0.8405 | 2.0 | 4544 | 0.4523 | 0.0234 | 0.8101 | 0.3434 | 0.8586 | 0.8435 | 0.7177 | 0.2700 | 0.8104 | 0.8130 | 0.9296 | 0.7802 | 0.9296 | 0.9421 | 0.9544 | 0.5954 |
0.6991 | 3.0 | 6816 | 0.5218 | 0.0214 | 0.8253 | 0.3595 | 0.8703 | 0.8563 | 0.7327 | 0.2829 | 0.8184 | 0.8238 | 0.9447 | 0.7721 | 0.9447 | 0.9534 | 0.9626 | 0.6190 |
0.3865 | 4.0 | 9088 | 0.8428 | 0.0155 | 0.8655 | 0.4279 | 0.8826 | 0.8808 | 0.8092 | 0.3453 | 0.8458 | 0.8667 | 0.9302 | 0.6992 | 0.9302 | 0.9417 | 0.9589 | 0.6917 |
0.1332 | 5.0 | 11360 | 0.9050 | 0.0147 | 0.8717 | 0.4368 | 0.8882 | 0.8857 | 0.8176 | 0.3560 | 0.8520 | 0.8728 | 0.9334 | 0.7011 | 0.9334 | 0.9438 | 0.9608 | 0.7014 |
Framework versions
- Transformers 4.28.0
- Pytorch 2.1.0+cu121
- Datasets 2.18.0
- Tokenizers 0.13.3
- Downloads last month
- 1