SETH_5e-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: 0.0716
- Precision: 0.7964
- Recall: 0.8036
- F1: 0.8000
- Accuracy: 0.9849
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 |
---|---|---|---|---|---|---|---|
0.3757 | 0.76 | 25 | 0.1924 | 0.0 | 0.0 | 0.0 | 0.9625 |
0.1119 | 1.52 | 50 | 0.0723 | 0.6237 | 0.7473 | 0.6799 | 0.9775 |
0.0565 | 2.27 | 75 | 0.0614 | 0.6569 | 0.7727 | 0.7101 | 0.9794 |
0.048 | 3.03 | 100 | 0.0586 | 0.6667 | 0.8655 | 0.7532 | 0.9801 |
0.0355 | 3.79 | 125 | 0.0519 | 0.7206 | 0.8345 | 0.7734 | 0.9835 |
0.0328 | 4.55 | 150 | 0.0532 | 0.7165 | 0.8455 | 0.7756 | 0.9831 |
0.0258 | 5.3 | 175 | 0.0539 | 0.7460 | 0.8382 | 0.7894 | 0.9835 |
0.022 | 6.06 | 200 | 0.0561 | 0.7612 | 0.7709 | 0.7660 | 0.9836 |
0.0189 | 6.82 | 225 | 0.0564 | 0.7636 | 0.74 | 0.7516 | 0.9828 |
0.0166 | 7.58 | 250 | 0.0597 | 0.7274 | 0.8491 | 0.7836 | 0.9836 |
0.0128 | 8.33 | 275 | 0.0626 | 0.8251 | 0.7636 | 0.7932 | 0.9854 |
0.0113 | 9.09 | 300 | 0.0603 | 0.8029 | 0.8 | 0.8015 | 0.9854 |
0.009 | 9.85 | 325 | 0.0687 | 0.8026 | 0.7909 | 0.7967 | 0.9857 |
0.0075 | 10.61 | 350 | 0.0716 | 0.7964 | 0.8036 | 0.8000 | 0.9849 |
Framework versions
- Transformers 4.27.4
- Pytorch 1.13.1+cu116
- Datasets 2.11.0
- Tokenizers 0.13.2
- Downloads last month
- 9