metadata
license: mit
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: Variome_0.0001_250
results: []
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: 0.0673
- Precision: 0.6558
- Recall: 0.5824
- F1: 0.6169
- 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 |
---|---|---|---|---|---|---|---|
0.4275 | 0.35 | 25 | 0.1841 | 0.0 | 0.0 | 0.0 | 0.9760 |
0.1851 | 0.69 | 50 | 0.1800 | 0.0 | 0.0 | 0.0 | 0.9760 |
0.1622 | 1.04 | 75 | 0.1746 | 0.0 | 0.0 | 0.0 | 0.9760 |
0.1617 | 1.39 | 100 | 0.1281 | 0.2462 | 0.0418 | 0.0715 | 0.9764 |
0.1185 | 1.74 | 125 | 0.1175 | 0.1430 | 0.1537 | 0.1481 | 0.9749 |
0.1104 | 2.08 | 150 | 0.1066 | 0.1982 | 0.1486 | 0.1698 | 0.9776 |
0.0906 | 2.43 | 175 | 0.1057 | 0.3385 | 0.2041 | 0.2547 | 0.9793 |
0.0922 | 2.78 | 200 | 0.0919 | 0.3749 | 0.3185 | 0.3444 | 0.9807 |
0.0801 | 3.12 | 225 | 0.0836 | 0.4769 | 0.4142 | 0.4433 | 0.9823 |
0.0655 | 3.47 | 250 | 0.0808 | 0.5989 | 0.4552 | 0.5172 | 0.9840 |
0.0606 | 3.82 | 275 | 0.0716 | 0.5641 | 0.5038 | 0.5323 | 0.9851 |
0.0485 | 4.17 | 300 | 0.0707 | 0.5975 | 0.5235 | 0.5580 | 0.9856 |
0.0464 | 4.51 | 325 | 0.0686 | 0.6370 | 0.5320 | 0.5798 | 0.9858 |
0.0423 | 4.86 | 350 | 0.0678 | 0.6343 | 0.5628 | 0.5964 | 0.9862 |
0.0387 | 5.21 | 375 | 0.0669 | 0.6523 | 0.5542 | 0.5993 | 0.9866 |
0.0336 | 5.56 | 400 | 0.0677 | 0.6573 | 0.5602 | 0.6049 | 0.9866 |
0.033 | 5.9 | 425 | 0.0673 | 0.6500 | 0.5568 | 0.5998 | 0.9866 |
0.031 | 6.25 | 450 | 0.0673 | 0.6477 | 0.5824 | 0.6133 | 0.9864 |
0.0276 | 6.6 | 475 | 0.0676 | 0.6573 | 0.5798 | 0.6162 | 0.9868 |
0.0255 | 6.94 | 500 | 0.0673 | 0.6558 | 0.5824 | 0.6169 | 0.9867 |
Framework versions
- Transformers 4.27.4
- Pytorch 1.13.1+cu116
- Datasets 2.11.0
- Tokenizers 0.13.2