<!-- 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_03_07
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: 0.0171
- Precision: 0.8141
- Recall: 0.8757
- F1: 0.8437
- Accuracy: 0.9969
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 |
|---|---|---|---|---|---|---|---|
| 0.501 | 1.47 | 25 | 0.1310 | 0.0 | 0.0 | 0.0 | 0.9786 |
| 0.096 | 2.94 | 50 | 0.0526 | 0.3789 | 0.3297 | 0.3526 | 0.9862 |
| 0.0436 | 4.41 | 75 | 0.0308 | 0.3918 | 0.5189 | 0.4465 | 0.9917 |
| 0.0301 | 5.88 | 100 | 0.0222 | 0.5853 | 0.6865 | 0.6318 | 0.9944 |
| 0.0144 | 7.35 | 125 | 0.0176 | 0.7246 | 0.8108 | 0.7653 | 0.9957 |
| 0.0091 | 8.82 | 150 | 0.0166 | 0.7536 | 0.8432 | 0.7959 | 0.9957 |
| 0.006 | 10.29 | 175 | 0.0143 | 0.8298 | 0.8432 | 0.8365 | 0.9968 |
| 0.0039 | 11.76 | 200 | 0.0154 | 0.8112 | 0.8595 | 0.8346 | 0.9966 |
| 0.0027 | 13.24 | 225 | 0.0172 | 0.7586 | 0.8324 | 0.7938 | 0.9963 |
| 0.0026 | 14.71 | 250 | 0.0164 | 0.8163 | 0.8649 | 0.8399 | 0.9966 |
| 0.002 | 16.18 | 275 | 0.0173 | 0.8 | 0.8649 | 0.8312 | 0.9968 |
| 0.0018 | 17.65 | 300 | 0.0185 | 0.8641 | 0.8595 | 0.8618 | 0.9965 |
| 0.0016 | 19.12 | 325 | 0.0175 | 0.7970 | 0.8703 | 0.8320 | 0.9968 |
| 0.0015 | 20.59 | 350 | 0.0171 | 0.8141 | 0.8757 | 0.8437 | 0.9969 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.13.1+cu116
- Datasets 2.10.1
- Tokenizers 0.13.2
