deberta-pii-owndlp-checkpoints

This model is a fine-tuned version of microsoft/deberta-v3-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 3.2718
• Precision: 0.0
• Recall: 0.0
• F1: 0.0
• Person Name F1: 0.0
• Contact Email F1: 0.0
• Contact Phone F1: 0.0
• Address F1: 0.0
• Organization F1: 0.0
• Credential Secret F1: 0.0
• Date Of Birth F1: 0.0
• Financial Bank Account F1: 0.0
• Gov Id Aadhaar F1: 0.0
• Gov Id Pan F1: 0.0
• Gov Id Gstin F1: 0.0
• Financial Ifsc F1: 0.0
• Job Title F1: 0.0
• Medical Condition F1: 0.0
• Username F1: 0.0
• Financial Credit Card F1: 0.0
• Financial Iban F1: 0.0
• Gov Id National Generic F1: 0.0
• Network Ip F1: 0.0

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-06
• train_batch_size: 8
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 4
• 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
• lr_scheduler_warmup_steps: 260
• num_epochs: 4

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Person Name F1	Contact Email F1	Contact Phone F1	Address F1	Organization F1	Credential Secret F1	Date Of Birth F1	Financial Bank Account F1	Gov Id Aadhaar F1	Gov Id Pan F1	Gov Id Gstin F1	Financial Ifsc F1	Job Title F1	Medical Condition F1	Username F1	Financial Credit Card F1	Financial Iban F1	Gov Id National Generic F1	Network Ip F1
13.6149	0.1538	100	2.0031	0.1247	0.0385	0.0589	0.1355	0.1572	0.0	0.0	0.0024	0.0	0.1015	0.0	0.0	0.0	0.0	0.0	0.0105	0.0861	0.0	0.0	0.0	0.0	0.0
3.7565	0.3077	200	0.0772	0.9811	0.9317	0.9557	0.9921	1.0	1.0	0.9971	0.9938	0.0333	1.0	0.9519	0.8116	0.9399	0.9945	0.9960	0.9712	1.0	0.9574	0.0	0.0	0.0	0.0
0.4267	0.4615	300	0.0397	0.9764	0.9859	0.9812	0.9943	0.9969	1.0	1.0	0.9801	0.8954	1.0	1.0	1.0	1.0	1.0	1.0	0.8665	1.0	0.9465	0.9962	0.0	0.0	0.0
0.8863	0.6154	400	0.1106	0.8581	0.9431	0.8986	0.8154	0.9937	0.9869	0.7898	0.9306	0.9658	0.9372	0.9836	0.9893	0.9622	0.7910	0.8848	0.9436	0.9925	0.88	0.9618	0.0	0.0	0.0
4.9823	0.7692	500	0.7189	0.5845	0.6033	0.5937	0.9124	0.8256	0.8631	0.2505	0.4516	0.0	0.9019	0.0	0.0279	0.3720	0.5150	0.8802	0.7924	0.8040	0.4823	0.0	0.0	0.0	0.0
10.0716	0.9231	600	3.2442	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
12.9110	1.0769	700	3.2475	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
12.4971	1.2308	800	3.2321	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
12.2021	1.3846	900	3.2752	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
11.8339	1.5385	1000	3.3241	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
11.7611	1.6923	1100	3.3389	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
11.3559	1.8462	1200	3.3377	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
11.3732	2.0	1300	3.2793	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
11.1335	2.1538	1400	3.2968	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
10.9102	2.3077	1500	3.2718	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0

Framework versions

• Transformers 5.16.1
• Pytorch 2.11.0+cu128
• Datasets 4.8.5
• Tokenizers 0.23.1