<!-- 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. -->
PubMedBERT-LitCovid-v1.1
This model is a fine-tuned version of microsoft/BiomedNLP-PubMedBERT-base-uncased-abstract on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.1198
- F1: 0.8985
- Roc Auc: 0.9368
- Accuracy: 0.7937
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
- num_epochs: 5
Training results
Training Loss | Epoch | Step | Validation Loss | F1 | Roc Auc | Accuracy |
---|---|---|---|---|---|---|
0.1187 | 1.0 | 1560 | 0.1141 | 0.8923 | 0.9335 | 0.7819 |
0.0976 | 2.0 | 3120 | 0.1063 | 0.8983 | 0.9325 | 0.7924 |
0.0702 | 3.0 | 4680 | 0.1147 | 0.8970 | 0.9420 | 0.7839 |
0.0534 | 4.0 | 6240 | 0.1198 | 0.8985 | 0.9368 | 0.7937 |
0.0391 | 5.0 | 7800 | 0.1266 | 0.8982 | 0.9384 | 0.7902 |
Framework versions
- Transformers 4.28.0
- Pytorch 2.0.0
- Datasets 2.1.0
- Tokenizers 0.13.3