Disease_Identification_SonatafyAI_BERT_v1
This model is a fine-tuned version of google-bert/bert-base-cased on the ncbi_disease dataset. It achieves the following results on the evaluation set:
- Loss: 0.1063
- Precision: 0.8247
- Recall: 0.8729
- F1: 0.8481
- Accuracy: 0.9840
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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 7
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.1208 | 1.0 | 680 | 0.0579 | 0.7339 | 0.8272 | 0.7778 | 0.9816 |
| 0.0438 | 2.0 | 1360 | 0.0616 | 0.7785 | 0.8755 | 0.8242 | 0.9836 |
| 0.0171 | 3.0 | 2040 | 0.0736 | 0.7949 | 0.8272 | 0.8107 | 0.9822 |
| 0.0096 | 4.0 | 2720 | 0.0857 | 0.8173 | 0.8640 | 0.8400 | 0.9844 |
| 0.0069 | 5.0 | 3400 | 0.0986 | 0.8050 | 0.8551 | 0.8293 | 0.9833 |
| 0.0016 | 6.0 | 4080 | 0.1055 | 0.8068 | 0.8755 | 0.8397 | 0.9834 |
| 0.0009 | 7.0 | 4760 | 0.1063 | 0.8247 | 0.8729 | 0.8481 | 0.9840 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2