tmvar_2e-05_250
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: 0.0128
- Precision: 0.8756
- Recall: 0.9135
- F1: 0.8942
- Accuracy: 0.9974
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: 500
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.486 | 1.0 | 25 | 0.0910 | 0.0 | 0.0 | 0.0 | 0.9858 |
| 0.0765 | 2.0 | 50 | 0.0410 | 0.6267 | 0.2541 | 0.3615 | 0.9889 |
| 0.0399 | 3.0 | 75 | 0.0230 | 0.6513 | 0.6865 | 0.6684 | 0.9941 |
| 0.0254 | 4.0 | 100 | 0.0176 | 0.7170 | 0.8216 | 0.7657 | 0.9957 |
| 0.0139 | 5.0 | 125 | 0.0129 | 0.8710 | 0.8757 | 0.8733 | 0.9968 |
| 0.0078 | 6.0 | 150 | 0.0107 | 0.9027 | 0.9027 | 0.9027 | 0.9974 |
| 0.0057 | 7.0 | 175 | 0.0110 | 0.8763 | 0.9189 | 0.8971 | 0.9975 |
| 0.0042 | 8.0 | 200 | 0.0113 | 0.8718 | 0.9189 | 0.8947 | 0.9971 |
| 0.003 | 9.0 | 225 | 0.0118 | 0.8802 | 0.9135 | 0.8966 | 0.9974 |
| 0.0022 | 10.0 | 250 | 0.0121 | 0.8877 | 0.8973 | 0.8925 | 0.9972 |
| 0.0019 | 11.0 | 275 | 0.0126 | 0.8756 | 0.9135 | 0.8942 | 0.9972 |
| 0.0016 | 12.0 | 300 | 0.0126 | 0.8802 | 0.9135 | 0.8966 | 0.9974 |
| 0.0015 | 13.0 | 325 | 0.0129 | 0.8769 | 0.9243 | 0.9 | 0.9974 |
| 0.0013 | 14.0 | 350 | 0.0128 | 0.8756 | 0.9135 | 0.8942 | 0.9974 |
Framework versions
- Transformers 4.27.4
- Pytorch 1.13.1+cu116
- Datasets 2.11.0
- Tokenizers 0.13.2