<!-- 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. -->
SETH_2e-05_0404_ES6
This model is a fine-tuned version of microsoft/BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0579
- Precision: 0.7801
- Recall: 0.8485
- F1: 0.8129
- Accuracy: 0.9863
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
- training_steps: 2000
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
0.5347 | 0.96 | 25 | 0.2086 | 0.0 | 0.0 | 0.0 | 0.9606 |
0.1372 | 1.92 | 50 | 0.0927 | 0.5702 | 0.7762 | 0.6574 | 0.9742 |
0.0648 | 2.88 | 75 | 0.0647 | 0.6081 | 0.8279 | 0.7012 | 0.9774 |
0.0494 | 3.85 | 100 | 0.0603 | 0.6395 | 0.8244 | 0.7203 | 0.9788 |
0.0448 | 4.81 | 125 | 0.0612 | 0.6106 | 0.8744 | 0.7190 | 0.9780 |
0.0334 | 5.77 | 150 | 0.0565 | 0.6564 | 0.8847 | 0.7537 | 0.9801 |
0.0293 | 6.73 | 175 | 0.0518 | 0.7465 | 0.8365 | 0.7890 | 0.9832 |
0.0224 | 7.69 | 200 | 0.0538 | 0.7010 | 0.8675 | 0.7754 | 0.9829 |
0.0186 | 8.65 | 225 | 0.0583 | 0.7508 | 0.8554 | 0.7997 | 0.9842 |
0.0174 | 9.62 | 250 | 0.0525 | 0.7604 | 0.8795 | 0.8156 | 0.9846 |
0.013 | 10.58 | 275 | 0.0642 | 0.7189 | 0.8847 | 0.7932 | 0.9839 |
0.0128 | 11.54 | 300 | 0.0572 | 0.7855 | 0.8571 | 0.8198 | 0.9863 |
0.0116 | 12.5 | 325 | 0.0579 | 0.7801 | 0.8485 | 0.8129 | 0.9863 |
Framework versions
- Transformers 4.27.4
- Pytorch 2.0.0+cu118
- Datasets 2.11.0
- Tokenizers 0.13.2