pii-dual-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: 0.0022
• Precision: 0.9963
• Recall: 0.9977
• F1: 0.9970
• Accuracy: 0.9995

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
0.0118	0.4638	2000	0.0101	0.9835	0.9912	0.9873	0.9973
0.0097	0.9275	4000	0.0049	0.9923	0.9960	0.9942	0.9988
0.0177	1.3912	6000	0.0030	0.9939	0.9957	0.9948	0.9990
0.0044	1.8550	8000	0.0030	0.9949	0.9967	0.9958	0.9991
0.0021	2.3186	10000	0.0026	0.9941	0.9957	0.9949	0.9992
0.0049	2.7824	12000	0.0018	0.9961	0.9978	0.9970	0.9994
0.0002	3.2460	14000	0.0023	0.9963	0.9979	0.9971	0.9994
0.0007	3.7098	16000	0.0022	0.9964	0.9978	0.9971	0.9995
0.0001	4.0	17252	0.0022	0.9963	0.9977	0.9970	0.9995

Framework versions

• Transformers 5.15.1
• Pytorch 2.13.0+cu130
• Datasets 5.0.1
• Tokenizers 0.22.2