<!-- 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_2e-05_0404_ES6_strict_tok
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: 0.0325
- Precision: 0.7972
- Recall: 0.8782
- F1: 0.8357
- Accuracy: 0.9909
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: 2000
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
0.5778 | 0.49 | 25 | 0.2034 | 0.0 | 0.0 | 0.0 | 0.9555 |
0.1384 | 0.98 | 50 | 0.0953 | 0.0 | 0.0 | 0.0 | 0.9705 |
0.0778 | 1.47 | 75 | 0.0841 | 0.0 | 0.0 | 0.0 | 0.9734 |
0.064 | 1.96 | 100 | 0.0506 | 0.6818 | 0.2284 | 0.3422 | 0.9827 |
0.0368 | 2.45 | 125 | 0.0424 | 0.6318 | 0.6447 | 0.6382 | 0.9882 |
0.0212 | 2.94 | 150 | 0.0360 | 0.7478 | 0.8579 | 0.7991 | 0.9899 |
0.0138 | 3.43 | 175 | 0.0398 | 0.7629 | 0.8985 | 0.8252 | 0.9899 |
0.013 | 3.92 | 200 | 0.0250 | 0.8502 | 0.8934 | 0.8713 | 0.9932 |
0.0079 | 4.41 | 225 | 0.0293 | 0.8579 | 0.8579 | 0.8579 | 0.9925 |
0.0055 | 4.9 | 250 | 0.0325 | 0.7972 | 0.8782 | 0.8357 | 0.9909 |
Framework versions
- Transformers 4.27.4
- Pytorch 2.0.0+cu118
- Datasets 2.11.0
- Tokenizers 0.13.3