<!-- 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. -->
tmvar
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: 0.0183
- Precision: 0.7718
- Recall: 0.8595
- F1: 0.8133
- Accuracy: 0.9963
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: 350
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.6322 | 1.47 | 25 | 0.1254 | 0.0 | 0.0 | 0.0 | 0.9786 |
| 0.083 | 2.94 | 50 | 0.0469 | 0.4729 | 0.5189 | 0.4948 | 0.9853 |
| 0.0408 | 4.41 | 75 | 0.0303 | 0.5409 | 0.6432 | 0.5877 | 0.9928 |
| 0.0277 | 5.88 | 100 | 0.0216 | 0.5890 | 0.6973 | 0.6386 | 0.9946 |
| 0.0128 | 7.35 | 125 | 0.0188 | 0.7427 | 0.8270 | 0.7826 | 0.9957 |
| 0.0087 | 8.82 | 150 | 0.0215 | 0.7051 | 0.8270 | 0.7612 | 0.9945 |
| 0.0058 | 10.29 | 175 | 0.0160 | 0.7744 | 0.8162 | 0.7947 | 0.9962 |
| 0.0044 | 11.76 | 200 | 0.0176 | 0.8010 | 0.8486 | 0.8241 | 0.9963 |
| 0.003 | 13.24 | 225 | 0.0175 | 0.7822 | 0.8541 | 0.8165 | 0.9966 |
| 0.0026 | 14.71 | 250 | 0.0182 | 0.7794 | 0.8595 | 0.8175 | 0.9964 |
| 0.0023 | 16.18 | 275 | 0.0179 | 0.7921 | 0.8649 | 0.8269 | 0.9968 |
| 0.002 | 17.65 | 300 | 0.0183 | 0.7745 | 0.8541 | 0.8123 | 0.9963 |
| 0.0019 | 19.12 | 325 | 0.0182 | 0.7756 | 0.8595 | 0.8154 | 0.9964 |
| 0.0019 | 20.59 | 350 | 0.0183 | 0.7718 | 0.8595 | 0.8133 | 0.9963 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.13.1+cu116
- Datasets 2.10.1
- Tokenizers 0.13.2
