<!-- 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_5e-05_0404_ES6
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: 0.0114
- Precision: 0.8311
- Recall: 0.9239
- F1: 0.875
- Accuracy: 0.9970
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: 2000
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
0.3577 | 0.49 | 25 | 0.1106 | 0.0 | 0.0 | 0.0 | 0.9822 |
0.0633 | 0.98 | 50 | 0.0529 | 0.2241 | 0.3299 | 0.2669 | 0.9824 |
0.037 | 1.47 | 75 | 0.0442 | 0.2006 | 0.3249 | 0.2481 | 0.9847 |
0.0343 | 1.96 | 100 | 0.0283 | 0.4245 | 0.5279 | 0.4706 | 0.9906 |
0.017 | 2.45 | 125 | 0.0216 | 0.6178 | 0.7056 | 0.6588 | 0.9934 |
0.0143 | 2.94 | 150 | 0.0171 | 0.6609 | 0.7817 | 0.7163 | 0.9943 |
0.0098 | 3.43 | 175 | 0.0151 | 0.6934 | 0.7462 | 0.7188 | 0.9950 |
0.0098 | 3.92 | 200 | 0.0091 | 0.8413 | 0.8883 | 0.8642 | 0.9975 |
0.0047 | 4.41 | 225 | 0.0131 | 0.7773 | 0.9036 | 0.8357 | 0.9962 |
0.0044 | 4.9 | 250 | 0.0109 | 0.7957 | 0.9289 | 0.8571 | 0.9966 |
0.0017 | 5.39 | 275 | 0.0108 | 0.8136 | 0.9086 | 0.8585 | 0.9970 |
0.0015 | 5.88 | 300 | 0.0098 | 0.875 | 0.9239 | 0.8988 | 0.9974 |
0.0013 | 6.37 | 325 | 0.0095 | 0.8824 | 0.9137 | 0.8978 | 0.9974 |
0.0008 | 6.86 | 350 | 0.0082 | 0.9109 | 0.9340 | 0.9223 | 0.9980 |
0.0013 | 7.35 | 375 | 0.0112 | 0.8663 | 0.8883 | 0.8772 | 0.9971 |
0.0007 | 7.84 | 400 | 0.0123 | 0.8683 | 0.9036 | 0.8856 | 0.9972 |
0.0008 | 8.33 | 425 | 0.0087 | 0.8867 | 0.9137 | 0.9 | 0.9976 |
0.0005 | 8.82 | 450 | 0.0095 | 0.885 | 0.8985 | 0.8917 | 0.9975 |
0.0004 | 9.31 | 475 | 0.0115 | 0.8436 | 0.9036 | 0.8725 | 0.9971 |
0.0004 | 9.8 | 500 | 0.0114 | 0.8311 | 0.9239 | 0.875 | 0.9970 |
Framework versions
- Transformers 4.27.4
- Pytorch 2.0.0+cu118
- Datasets 2.11.0
- Tokenizers 0.13.2