distilbert-base-uncased-three
This model is a fine-tuned version of devtibo/distilbert-base-uncased-pii-200 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1168
- Overall Precision: 0.6825
- Overall Recall: 0.7213
- Overall F1: 0.7014
- Overall Accuracy: 0.9656
- Accountname F1: 0.9839
- Accountnumber F1: 0.9919
- Age F1: 0.9055
- Amount F1: 0.9706
- Bic F1: 0.9154
- Bitcoinaddress F1: 0.9585
- Buildingnumber F1: 0.6502
- City F1: 0.5062
- Companyname F1: 0.7796
- County F1: 0.9751
- Creditcardcvv F1: 0.7854
- Creditcardissuer F1: 0.9774
- Creditcardnumber F1: 0.8696
- Currency F1: 0.7022
- Currencycode F1: 0.8587
- Currencyname F1: 0.3934
- Currencysymbol F1: 0.9382
- Date F1: 0.7854
- Dob F1: 0.5582
- Email F1: 0.6560
- Ethereumaddress F1: 0.9908
- Eyecolor F1: 0.9558
- Firstname F1: 0.9398
- Gender F1: 0.9786
- Height F1: 0.9885
- Iban F1: 0.9554
- Ip F1: 0.3429
- Ipv4 F1: 0.8633
- Ipv6 F1: 0.7580
- Jobarea F1: 0.9149
- Jobtitle F1: 0.6491
- Jobtype F1: 0.9315
- Lastname F1: 0.4773
- Litecoinaddress F1: 0.8642
- Mac F1: 0.9905
- Maskednumber F1: 0.8479
- Middlename F1: 0.9547
- Nearbygpscoordinate F1: 0.6901
- Ordinaldirection F1: 0.9885
- Password F1: 0.5102
- Phoneimei F1: 0.9874
- Phonenumber F1: 0.6291
- Pin F1: 0.8176
- Prefix F1: 0.9280
- Secondaryaddress F1: 0.5806
- Sex F1: 0.6462
- Ssn F1: 0.5140
- State F1: 0.5714
- Street F1: 0.7272
- Time F1: 0.5874
- Url F1: 0.9842
- Useragent F1: 0.9875
- Username F1: 0.7342
- Vehiclevin F1: 0.9890
- Vehiclevrm F1: 0.9797
- Zipcode F1: 0.5130
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: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.2
- num_epochs: 7
Training results
Training Loss | Epoch | Step | Validation Loss | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy | Accountname F1 | Accountnumber F1 | Age F1 | Amount F1 | Bic F1 | Bitcoinaddress F1 | Buildingnumber F1 | City F1 | Companyname F1 | County F1 | Creditcardcvv F1 | Creditcardissuer F1 | Creditcardnumber F1 | Currency F1 | Currencycode F1 | Currencyname F1 | Currencysymbol F1 | Date F1 | Dob F1 | Email F1 | Ethereumaddress F1 | Eyecolor F1 | Firstname F1 | Gender F1 | Height F1 | Iban F1 | Ip F1 | Ipv4 F1 | Ipv6 F1 | Jobarea F1 | Jobtitle F1 | Jobtype F1 | Lastname F1 | Litecoinaddress F1 | Mac F1 | Maskednumber F1 | Middlename F1 | Nearbygpscoordinate F1 | Ordinaldirection F1 | Password F1 | Phoneimei F1 | Phonenumber F1 | Pin F1 | Prefix F1 | Secondaryaddress F1 | Sex F1 | Ssn F1 | State F1 | Street F1 | Time F1 | Url F1 | Useragent F1 | Username F1 | Vehiclevin F1 | Vehiclevrm F1 | Zipcode F1 |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
0.1616 | 1.0 | 2597 | 0.1596 | 0.5945 | 0.5984 | 0.5965 | 0.9491 | 0.9877 | 0.9819 | 0.8851 | 0.9706 | 0.7989 | 0.9448 | 0.4507 | 0.3387 | 0.7151 | 0.9636 | 0.8061 | 0.9805 | 0.8387 | 0.7868 | 0.8249 | 0.1792 | 0.9326 | 0.6860 | 0.3981 | 0.5203 | 1.0 | 0.9474 | 0.9176 | 0.9643 | 0.9857 | 0.8787 | 0.1314 | 0.7297 | 0.6964 | 0.9059 | 0.5193 | 0.9343 | 0.3508 | 0.8688 | 0.9882 | 0.8551 | 0.9444 | 0.6050 | 0.9771 | 0.3679 | 0.9789 | 0.5074 | 0.8469 | 0.9266 | 0.4529 | 0.4990 | 0.3068 | 0.4534 | 0.5742 | 0.3881 | 0.9777 | 0.9943 | 0.6757 | 0.9699 | 0.9669 | 0.3959 |
0.1182 | 2.0 | 5194 | 0.1270 | 0.6434 | 0.6620 | 0.6526 | 0.9601 | 0.9886 | 0.9910 | 0.9110 | 0.9713 | 0.9067 | 0.9516 | 0.5231 | 0.4309 | 0.7426 | 0.9731 | 0.8083 | 0.9821 | 0.8616 | 0.6374 | 0.8282 | 0.3942 | 0.9291 | 0.7471 | 0.4915 | 0.6238 | 1.0 | 0.9497 | 0.9300 | 0.9679 | 0.9744 | 0.9256 | 0.2722 | 0.7812 | 0.7222 | 0.9046 | 0.5927 | 0.9273 | 0.4057 | 0.8408 | 0.9834 | 0.8373 | 0.9537 | 0.6369 | 0.9860 | 0.4106 | 0.9924 | 0.5629 | 0.8453 | 0.9331 | 0.5238 | 0.5843 | 0.4167 | 0.5333 | 0.6620 | 0.4942 | 0.9859 | 0.9784 | 0.7232 | 0.9945 | 0.9847 | 0.4792 |
0.095 | 3.0 | 7791 | 0.1190 | 0.6799 | 0.7027 | 0.6911 | 0.9633 | 0.9849 | 0.9883 | 0.9089 | 0.9778 | 0.9175 | 0.9565 | 0.6166 | 0.4787 | 0.7805 | 0.9732 | 0.7961 | 0.9726 | 0.8591 | 0.7141 | 0.816 | 0.3265 | 0.9327 | 0.7712 | 0.5464 | 0.6556 | 0.9954 | 0.9135 | 0.9357 | 0.9757 | 0.9885 | 0.9495 | 0.3161 | 0.7982 | 0.7324 | 0.9223 | 0.6346 | 0.9329 | 0.4629 | 0.8812 | 0.9882 | 0.8618 | 0.9550 | 0.6813 | 0.9860 | 0.4634 | 0.9850 | 0.6165 | 0.8771 | 0.9313 | 0.5760 | 0.6183 | 0.5156 | 0.5520 | 0.6985 | 0.5628 | 0.9906 | 0.9909 | 0.7354 | 0.9781 | 0.9823 | 0.5061 |
0.0734 | 4.0 | 10388 | 0.1168 | 0.6825 | 0.7213 | 0.7014 | 0.9656 | 0.9839 | 0.9919 | 0.9055 | 0.9706 | 0.9154 | 0.9585 | 0.6502 | 0.5062 | 0.7796 | 0.9751 | 0.7854 | 0.9774 | 0.8696 | 0.7022 | 0.8587 | 0.3934 | 0.9382 | 0.7854 | 0.5582 | 0.6560 | 0.9908 | 0.9558 | 0.9398 | 0.9786 | 0.9885 | 0.9554 | 0.3429 | 0.8633 | 0.7580 | 0.9149 | 0.6491 | 0.9315 | 0.4773 | 0.8642 | 0.9905 | 0.8479 | 0.9547 | 0.6901 | 0.9885 | 0.5102 | 0.9874 | 0.6291 | 0.8176 | 0.9280 | 0.5806 | 0.6462 | 0.5140 | 0.5714 | 0.7272 | 0.5874 | 0.9842 | 0.9875 | 0.7342 | 0.9890 | 0.9797 | 0.5130 |
0.059 | 5.0 | 12985 | 0.1288 | 0.7020 | 0.7296 | 0.7155 | 0.9676 | 0.9877 | 0.9964 | 0.9112 | 0.9755 | 0.8916 | 0.9630 | 0.6625 | 0.5331 | 0.7988 | 0.9795 | 0.7875 | 0.9774 | 0.8723 | 0.7609 | 0.8470 | 0.3713 | 0.9341 | 0.7870 | 0.5871 | 0.6846 | 0.9954 | 0.9529 | 0.9383 | 0.9796 | 0.9856 | 0.9448 | 0.3790 | 0.8461 | 0.7774 | 0.9153 | 0.6635 | 0.9330 | 0.4944 | 0.8840 | 0.9929 | 0.8605 | 0.9504 | 0.7029 | 0.9884 | 0.4850 | 0.9862 | 0.6502 | 0.8463 | 0.9286 | 0.5908 | 0.6563 | 0.5658 | 0.5805 | 0.7327 | 0.6162 | 0.9906 | 0.9920 | 0.7565 | 0.9834 | 0.9897 | 0.5355 |
0.0485 | 6.0 | 15582 | 0.1354 | 0.7096 | 0.7368 | 0.7229 | 0.9683 | 0.9858 | 0.9910 | 0.9081 | 0.9705 | 0.9222 | 0.9587 | 0.6713 | 0.5425 | 0.8042 | 0.9768 | 0.8231 | 0.9806 | 0.8703 | 0.7409 | 0.8580 | 0.3728 | 0.9370 | 0.7969 | 0.6113 | 0.6870 | 0.9954 | 0.9584 | 0.9420 | 0.9777 | 0.9856 | 0.9545 | 0.4193 | 0.8126 | 0.7457 | 0.9164 | 0.6659 | 0.9343 | 0.5057 | 0.8723 | 0.9952 | 0.8629 | 0.9576 | 0.7208 | 0.9860 | 0.5212 | 0.9862 | 0.6660 | 0.8581 | 0.9281 | 0.6021 | 0.6682 | 0.5675 | 0.6020 | 0.7395 | 0.6280 | 0.9915 | 0.9898 | 0.7579 | 0.9835 | 0.9872 | 0.5406 |
0.0419 | 7.0 | 18179 | 0.1417 | 0.7091 | 0.7375 | 0.7231 | 0.9688 | 0.9886 | 0.9919 | 0.9061 | 0.9690 | 0.9350 | 0.9575 | 0.6710 | 0.5395 | 0.8004 | 0.9777 | 0.8239 | 0.9838 | 0.8692 | 0.7638 | 0.8678 | 0.3721 | 0.9352 | 0.7961 | 0.6148 | 0.6874 | 0.9985 | 0.9582 | 0.9421 | 0.9776 | 0.9856 | 0.9531 | 0.4076 | 0.8189 | 0.7532 | 0.9158 | 0.6685 | 0.9342 | 0.5019 | 0.8642 | 0.9929 | 0.8599 | 0.9562 | 0.7162 | 0.9860 | 0.5209 | 0.9837 | 0.6680 | 0.8506 | 0.9218 | 0.6089 | 0.6643 | 0.5589 | 0.6056 | 0.7424 | 0.6303 | 0.9915 | 0.9886 | 0.7623 | 0.9890 | 0.9897 | 0.5558 |
Framework versions
- Transformers 4.44.0
- Pytorch 2.4.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1
- Downloads last month
- 6
Model tree for modeldev/distilbert-base-uncased-three
Base model
distilbert/distilbert-base-uncased
Finetuned
modeldev/distilbert-base-uncased-pii-200