my_pii_model
This model is a fine-tuned version of distilbert-base-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0010
Model description
num_labels=13
id2label = { 0: 'I-STREET_ADDRESS', 1: 'I-ID_NUM', 2: 'I-NAME_STUDENT', 3: 'B-PHONE_NUM', 4: 'B-USERNAME', 5: 'B-EMAIL', 6: 'I-PHONE_NUM', 7: 'O', 8: 'B-STREET_ADDRESS', 9: 'B-URL_PERSONAL', 10: 'I-URL_PERSONAL', 11: 'B-ID_NUM', 12: 'B-NAME_STUDENT' }
label2id = { 'I-STREET_ADDRESS': 0, 'I-ID_NUM': 1, 'I-NAME_STUDENT': 2, 'B-PHONE_NUM': 3, 'B-USERNAME': 4, 'B-EMAIL': 5, 'I-PHONE_NUM': 6, 'O': 7, 'B-STREET_ADDRESS': 8, 'B-URL_PERSONAL': 9, 'I-URL_PERSONAL': 10, 'B-ID_NUM': 11, 'B-NAME_STUDENT': 12 }
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
- num_epochs: 3
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
No log | 1.0 | 383 | 0.0016 |
0.04 | 2.0 | 766 | 0.0010 |
0.0012 | 3.0 | 1149 | 0.0010 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2
- Downloads last month
- 3