Variome_0.0001_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: 0.0638
- Precision: 0.6586
- Recall: 0.5816
- F1: 0.6177
- Accuracy: 0.9867
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: 0.0001
- 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 |
---|---|---|---|---|---|---|---|
0.3778 | 0.35 | 25 | 0.1802 | 0.0 | 0.0 | 0.0 | 0.9760 |
0.1563 | 0.69 | 50 | 0.1200 | 0.4524 | 0.0162 | 0.0313 | 0.9763 |
0.1061 | 1.04 | 75 | 0.1041 | 0.3604 | 0.2767 | 0.3130 | 0.9799 |
0.0981 | 1.39 | 100 | 0.0902 | 0.4585 | 0.3826 | 0.4171 | 0.9814 |
0.0807 | 1.74 | 125 | 0.0783 | 0.5129 | 0.4244 | 0.4645 | 0.9835 |
0.0731 | 2.08 | 150 | 0.0727 | 0.5513 | 0.5047 | 0.5270 | 0.9844 |
0.0526 | 2.43 | 175 | 0.0720 | 0.6368 | 0.5167 | 0.5705 | 0.9856 |
0.0604 | 2.78 | 200 | 0.0686 | 0.589 | 0.5030 | 0.5426 | 0.9849 |
0.0542 | 3.12 | 225 | 0.0671 | 0.6131 | 0.5371 | 0.5726 | 0.9856 |
0.0441 | 3.47 | 250 | 0.0669 | 0.6635 | 0.5389 | 0.5947 | 0.9860 |
0.0438 | 3.82 | 275 | 0.0667 | 0.625 | 0.5423 | 0.5807 | 0.9859 |
0.0381 | 4.17 | 300 | 0.0658 | 0.6562 | 0.5525 | 0.5999 | 0.9858 |
0.0404 | 4.51 | 325 | 0.0648 | 0.6578 | 0.5713 | 0.6115 | 0.9862 |
0.0341 | 4.86 | 350 | 0.0625 | 0.6637 | 0.5679 | 0.6121 | 0.9865 |
0.0298 | 5.21 | 375 | 0.0646 | 0.6727 | 0.5739 | 0.6194 | 0.9868 |
0.029 | 5.56 | 400 | 0.0643 | 0.6569 | 0.5739 | 0.6126 | 0.9861 |
0.0287 | 5.9 | 425 | 0.0637 | 0.6713 | 0.5739 | 0.6188 | 0.9869 |
0.027 | 6.25 | 450 | 0.0637 | 0.6660 | 0.5739 | 0.6165 | 0.9868 |
0.0236 | 6.6 | 475 | 0.0639 | 0.6644 | 0.5833 | 0.6212 | 0.9869 |
0.0233 | 6.94 | 500 | 0.0638 | 0.6586 | 0.5816 | 0.6177 | 0.9867 |
Framework versions
- Transformers 4.27.4
- Pytorch 1.13.1+cu116
- Datasets 2.11.0
- Tokenizers 0.13.2
- Downloads last month
- 2
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.