pii-dual-nym-mmbert-base-v2
This model is a fine-tuned version of jhu-clsp/mmBERT-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0020
- Precision: 0.9974
- Recall: 0.9982
- F1: 0.9978
- Accuracy: 0.9996
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: 32
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine_with_restarts
- lr_scheduler_warmup_steps: 200
- num_epochs: 4
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.0143 | 0.8873 | 2000 | 0.0044 | 0.9934 | 0.9956 | 0.9945 | 0.9989 |
| 0.0048 | 1.7746 | 4000 | 0.0030 | 0.9961 | 0.9970 | 0.9966 | 0.9993 |
| 0.0008 | 2.6619 | 6000 | 0.0022 | 0.9971 | 0.9978 | 0.9975 | 0.9995 |
| 0.0000 | 3.5492 | 8000 | 0.0020 | 0.9973 | 0.9981 | 0.9977 | 0.9996 |
| 0.0002 | 4.0 | 9016 | 0.0020 | 0.9974 | 0.9982 | 0.9978 | 0.9996 |
Framework versions
- Transformers 5.18.0
- Pytorch 2.14.1+cu130
- Datasets 5.0.1
- Tokenizers 0.23.2