deberta-v3-base_finetuned_bluegennx_run2
This model is a fine-tuned version of microsoft/deberta-v3-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0737
- Overall Precision: 0.7273
- Overall Recall: 0.7428
- Overall F1: 0.7350
- Overall Accuracy: 0.9752
- Aadhar F1: 0.8128
- Age F1: 0.4700
- City F1: 0.7686
- Country F1: 0.7226
- Creditcardcvv F1: 0.7531
- Creditcardnumber F1: 0.8109
- Date F1: 0.7126
- Dateofbirth F1: 0.7262
- Email F1: 0.6935
- Expiry F1: 0.6621
- Organization F1: 0.7623
- Pan F1: 0.7772
- Person F1: 0.7568
- Phonenumber F1: 0.8194
- Secondary F1: 0.6278
- State F1: 0.7735
- Time F1: 0.7856
- Url F1: 0.5824
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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine_with_restarts
- lr_scheduler_warmup_ratio: 0.2
- num_epochs: 5
Training results
Training Loss | Epoch | Step | Validation Loss | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy | Aadhar F1 | Age F1 | City F1 | Country F1 | Creditcardcvv F1 | Creditcardnumber F1 | Date F1 | Dateofbirth F1 | Email F1 | Expiry F1 | Organization F1 | Pan F1 | Person F1 | Phonenumber F1 | Secondary F1 | State F1 | Time F1 | Url F1 |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
0.1576 | 1.0 | 3893 | 0.1289 | 0.5166 | 0.5445 | 0.5302 | 0.9559 | 0.6073 | 0.1745 | 0.5790 | 0.5463 | 0.5707 | 0.6816 | 0.4834 | 0.4489 | 0.4808 | 0.5009 | 0.6085 | 0.5667 | 0.5383 | 0.5811 | 0.4273 | 0.6592 | 0.5824 | 0.2314 |
0.1075 | 2.0 | 7786 | 0.1151 | 0.5991 | 0.6001 | 0.5996 | 0.9610 | 0.7012 | 0.2439 | 0.6649 | 0.5689 | 0.6735 | 0.6950 | 0.5229 | 0.6065 | 0.5176 | 0.4904 | 0.6910 | 0.7248 | 0.5493 | 0.6810 | 0.5406 | 0.6382 | 0.6816 | 0.3492 |
0.0804 | 3.0 | 11679 | 0.0841 | 0.6783 | 0.7045 | 0.6911 | 0.9709 | 0.7826 | 0.3554 | 0.7372 | 0.6909 | 0.7276 | 0.7621 | 0.6459 | 0.7272 | 0.6303 | 0.6235 | 0.7329 | 0.7324 | 0.6816 | 0.7855 | 0.5912 | 0.7620 | 0.7529 | 0.4652 |
0.0532 | 4.0 | 15572 | 0.0737 | 0.7273 | 0.7428 | 0.7350 | 0.9752 | 0.8128 | 0.4700 | 0.7686 | 0.7226 | 0.7531 | 0.8109 | 0.7126 | 0.7262 | 0.6935 | 0.6621 | 0.7623 | 0.7772 | 0.7568 | 0.8194 | 0.6278 | 0.7735 | 0.7856 | 0.5824 |
0.0381 | 5.0 | 19465 | 0.0753 | 0.7372 | 0.7589 | 0.7479 | 0.9768 | 0.8278 | 0.4925 | 0.7705 | 0.7185 | 0.7832 | 0.8258 | 0.7231 | 0.7605 | 0.7027 | 0.6676 | 0.7700 | 0.8011 | 0.7591 | 0.8305 | 0.6558 | 0.7828 | 0.7978 | 0.6144 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.1.0+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2
- Downloads last month
- 6
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.
Model tree for C4Scale/deberta-v3-base_finetuned_bluegennx_run2
Base model
microsoft/deberta-v3-base