SETH_5e-5_0.01_29_03
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: 0.0773
- Precision: 0.6657
- Recall: 0.8055
- F1: 0.7290
- Accuracy: 0.9783
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: 5e-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 |
---|---|---|---|---|---|---|---|
0.363 | 0.96 | 25 | 0.1388 | 0.0 | 0.0 | 0.0 | 0.9583 |
0.0975 | 1.92 | 50 | 0.0746 | 0.5652 | 0.6867 | 0.6200 | 0.9724 |
0.0581 | 2.88 | 75 | 0.0613 | 0.6667 | 0.7952 | 0.7253 | 0.9768 |
0.045 | 3.85 | 100 | 0.0606 | 0.6313 | 0.8606 | 0.7283 | 0.9775 |
0.0396 | 4.81 | 125 | 0.0717 | 0.6058 | 0.8675 | 0.7134 | 0.9743 |
0.0296 | 5.77 | 150 | 0.0667 | 0.6307 | 0.8554 | 0.7261 | 0.9771 |
0.0237 | 6.73 | 175 | 0.0665 | 0.7086 | 0.8537 | 0.7744 | 0.9809 |
0.0191 | 7.69 | 200 | 0.0633 | 0.6908 | 0.8537 | 0.7637 | 0.9800 |
0.0133 | 8.65 | 225 | 0.0751 | 0.7878 | 0.7986 | 0.7932 | 0.9834 |
0.0137 | 9.62 | 250 | 0.0773 | 0.6657 | 0.8055 | 0.7290 | 0.9783 |
Framework versions
- Transformers 4.27.3
- Pytorch 1.13.1+cu116
- Datasets 2.10.1
- Tokenizers 0.13.2
- Downloads last month
- 3