<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
distilbert_finetuned_ai4privacy_50k
This model is a fine-tuned version of distilbert-base-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0391
- Overall Precision: 0.9187
- Overall Recall: 0.9520
- Overall F1: 0.9351
- Overall Accuracy: 0.9856
- Accountname F1: 0.9953
- Accountnumber F1: 0.9991
- Age F1: 0.9452
- Amount F1: 0.9964
- Bic F1: 0.9915
- Bitcoinaddress F1: 0.9992
- Buildingnumber F1: 0.9828
- City F1: 0.9833
- Companyname F1: 0.8652
- Company Name F1: 0.3746
- County F1: 0.9915
- Creditcardcvv F1: 0.9519
- Creditcardissuer F1: 0.9922
- Creditcardnumber F1: 0.9817
- Currency F1: 0.7452
- Currencycode F1: 0.8127
- Currencyname F1: 0.1975
- Currencysymbol F1: 0.8858
- Date F1: 0.9027
- Dob F1: 0.7677
- Email F1: 0.9994
- Ethereumaddress F1: 1.0
- Eyecolor F1: 0.9789
- Firstname F1: 0.9812
- Fullname F1: 0.0
- Gender F1: 0.9861
- Height F1: 1.0
- Iban F1: 0.9920
- Ip F1: 0.2044
- Ipv4 F1: 0.7792
- Ipv6 F1: 0.7179
- Jobarea F1: 0.9660
- Jobdescriptor F1: 0.4174
- Jobtitle F1: 0.9661
- Jobtype F1: 0.9803
- Lastname F1: 0.9774
- Litecoinaddress F1: 0.9979
- Mac F1: 1.0
- Maskednumber F1: 0.9757
- Middlename F1: 0.9216
- Nearbygpscoordinate F1: 1.0
- Ordinaldirection F1: 0.9812
- Password F1: 1.0
- Phoneimei F1: 1.0
- Phonenumber F1: 0.9127
- Phone Number F1: 0.4689
- Pin F1: 0.9377
- Prefix F1: 0.9531
- Secondaryaddress F1: 0.9973
- Sex F1: 0.9872
- Ssn F1: 0.9968
- State F1: 0.9929
- Street F1: 0.9937
- Streetaddress F1: 0.9636
- Suffix F1: 0.8958
- Time F1: 0.9855
- Url F1: 1.0
- Useragent F1: 1.0
- Username F1: 0.9977
- Vehiclevin F1: 1.0
- Vehiclevrm F1: 1.0
- Zipcode F1: 0.9908
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 | Company Name 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 | Fullname F1 | Gender F1 | Height F1 | Iban F1 | Ip F1 | Ipv4 F1 | Ipv6 F1 | Jobarea F1 | Jobdescriptor 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 | Phone Number F1 | Pin F1 | Prefix F1 | Secondaryaddress F1 | Sex F1 | Ssn F1 | State F1 | Street F1 | Streetaddress F1 | Suffix F1 | Time F1 | Url F1 | Useragent F1 | Username F1 | Vehiclevin F1 | Vehiclevrm F1 | Zipcode F1 |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
No log | 1.0 | 364 | 0.5224 | 0.3857 | 0.4077 | 0.3963 | 0.8528 | 0.0036 | 0.4011 | 0.0814 | 0.1482 | 0.2141 | 0.6117 | 0.0 | 0.2425 | 0.3185 | 0.0 | 0.0 | 0.0 | 0.0 | 0.2641 | 0.0962 | 0.0 | 0.0 | 0.0093 | 0.7196 | 0.0 | 0.8072 | 0.7088 | 0.0 | 0.5038 | 0.0 | 0.0 | 0.0 | 0.5131 | 0.0 | 0.7460 | 0.692 | 0.0 | 0.0 | 0.0321 | 0.0 | 0.2467 | 0.0190 | 0.8748 | 0.0762 | 0.0 | 0.9511 | 0.0 | 0.7917 | 0.8912 | 0.6407 | 0.0 | 0.0 | 0.0350 | 0.2806 | 0.0034 | 0.1091 | 0.0244 | 0.025 | 0.0 | 0.0 | 0.7702 | 0.9302 | 0.9129 | 0.2087 | 0.7389 | 0.4568 | 0.0596 |
1.4767 | 2.0 | 728 | 0.1843 | 0.7151 | 0.7894 | 0.7504 | 0.9370 | 0.9537 | 0.9528 | 0.8139 | 0.6502 | 0.3093 | 0.8966 | 0.5800 | 0.7681 | 0.8101 | 0.0 | 0.6801 | 0.3333 | 0.8204 | 0.5302 | 0.5135 | 0.0698 | 0.0 | 0.4230 | 0.7757 | 0.0105 | 0.9944 | 0.9866 | 0.6159 | 0.8626 | 0.0 | 0.8233 | 0.9184 | 0.9336 | 0.0 | 0.8268 | 0.5211 | 0.4871 | 0.0 | 0.7736 | 0.7565 | 0.7796 | 0.5249 | 0.9785 | 0.6228 | 0.4557 | 1.0 | 0.9349 | 0.8440 | 0.9947 | 0.8196 | 0.0 | 0.0131 | 0.868 | 0.9533 | 0.9479 | 0.8376 | 0.6811 | 0.6805 | 0.0 | 0.0 | 0.9374 | 0.9893 | 0.9878 | 0.9507 | 0.8353 | 0.6723 | 0.6823 |
0.234 | 3.0 | 1092 | 0.1177 | 0.7948 | 0.8772 | 0.8340 | 0.9505 | 0.9841 | 0.9853 | 0.8982 | 0.9275 | 0.9183 | 0.9594 | 0.7094 | 0.9091 | 0.8644 | 0.0 | 0.9315 | 0.7878 | 0.9630 | 0.8699 | 0.6722 | 0.3686 | 0.0 | 0.6688 | 0.8226 | 0.3718 | 0.9972 | 0.9867 | 0.9235 | 0.9205 | 0.0 | 0.9358 | 0.9780 | 0.9675 | 0.0095 | 0.8273 | 0.2704 | 0.8628 | 0.0 | 0.9069 | 0.8922 | 0.8769 | 0.8371 | 0.9900 | 0.8000 | 0.6298 | 1.0 | 0.9530 | 0.9687 | 1.0 | 0.8856 | 0.0 | 0.5096 | 0.9064 | 0.9892 | 0.9704 | 0.9813 | 0.8647 | 0.8294 | 0.0 | 0.0396 | 0.9567 | 0.9877 | 0.9921 | 0.9885 | 0.9713 | 0.9623 | 0.8410 |
0.234 | 4.0 | 1456 | 0.0896 | 0.8906 | 0.9192 | 0.9046 | 0.9625 | 0.9757 | 0.9870 | 0.9038 | 0.9761 | 0.9802 | 0.9785 | 0.8564 | 0.9669 | 0.8456 | 0.0 | 0.9584 | 0.8723 | 0.9752 | 0.9314 | 0.7429 | 0.6995 | 0.0 | 0.7888 | 0.8410 | 0.5058 | 0.9994 | 0.9955 | 0.9378 | 0.9418 | 0.0 | 0.9586 | 0.9852 | 0.9818 | 0.0 | 0.8300 | 0.8198 | 0.9329 | 0.0 | 0.9471 | 0.9469 | 0.9041 | 0.9325 | 0.9943 | 0.8935 | 0.7718 | 1.0 | 0.9728 | 0.9976 | 1.0 | 0.8801 | 0.0 | 0.8012 | 0.9072 | 0.9955 | 0.9746 | 0.9885 | 0.9903 | 0.9368 | 0.2105 | 0.5430 | 0.9668 | 0.9969 | 0.9939 | 0.9931 | 1.0 | 0.9957 | 0.9307 |
0.1256 | 5.0 | 1820 | 0.0676 | 0.9169 | 0.9380 | 0.9273 | 0.9668 | 0.9887 | 0.9895 | 0.9423 | 0.9956 | 1.0 | 0.9895 | 0.9231 | 0.9790 | 0.8594 | 0.0 | 0.9906 | 0.9178 | 0.9813 | 0.9453 | 0.7479 | 0.7579 | 0.0504 | 0.8594 | 0.8790 | 0.6790 | 1.0 | 0.9978 | 0.9737 | 0.9684 | 0.0 | 0.9753 | 0.9951 | 0.9897 | 0.0 | 0.8287 | 0.8095 | 0.9485 | 0.0645 | 0.9552 | 0.9619 | 0.9682 | 0.9811 | 1.0 | 0.9187 | 0.8607 | 1.0 | 0.9778 | 0.9968 | 1.0 | 0.8818 | 0.2353 | 0.8723 | 0.9463 | 0.9973 | 0.9821 | 0.9958 | 0.9920 | 0.9702 | 0.6026 | 0.7978 | 0.9791 | 1.0 | 0.9991 | 0.9977 | 1.0 | 1.0 | 0.9665 |
0.0873 | 6.0 | 2184 | 0.0505 | 0.9128 | 0.9479 | 0.9300 | 0.9806 | 0.9906 | 0.9965 | 0.9432 | 0.9964 | 0.9915 | 0.9880 | 0.9802 | 0.9798 | 0.8663 | 0.0724 | 0.9897 | 0.9437 | 0.9922 | 0.9842 | 0.7615 | 0.8011 | 0.1120 | 0.8843 | 0.8978 | 0.7372 | 1.0 | 1.0 | 0.9815 | 0.9777 | 0.0 | 0.9829 | 1.0 | 0.9920 | 0.1714 | 0.8154 | 0.6633 | 0.9653 | 0.2157 | 0.9581 | 0.9787 | 0.9740 | 0.9645 | 0.9986 | 0.9773 | 0.8994 | 1.0 | 0.9795 | 0.9992 | 1.0 | 0.8959 | 0.4000 | 0.9157 | 0.9475 | 0.9973 | 0.9846 | 0.9979 | 0.9912 | 0.9866 | 0.9550 | 0.8737 | 0.9800 | 1.0 | 1.0 | 0.9985 | 1.0 | 1.0 | 0.9852 |
0.0605 | 7.0 | 2548 | 0.0391 | 0.9187 | 0.9520 | 0.9351 | 0.9856 | 0.9953 | 0.9991 | 0.9452 | 0.9964 | 0.9915 | 0.9992 | 0.9828 | 0.9833 | 0.8652 | 0.3746 | 0.9915 | 0.9519 | 0.9922 | 0.9817 | 0.7452 | 0.8127 | 0.1975 | 0.8858 | 0.9027 | 0.7677 | 0.9994 | 1.0 | 0.9789 | 0.9812 | 0.0 | 0.9861 | 1.0 | 0.9920 | 0.2044 | 0.7792 | 0.7179 | 0.9660 | 0.4174 | 0.9661 | 0.9803 | 0.9774 | 0.9979 | 1.0 | 0.9757 | 0.9216 | 1.0 | 0.9812 | 1.0 | 1.0 | 0.9127 | 0.4689 | 0.9377 | 0.9531 | 0.9973 | 0.9872 | 0.9968 | 0.9929 | 0.9937 | 0.9636 | 0.8958 | 0.9855 | 1.0 | 1.0 | 0.9977 | 1.0 | 1.0 | 0.9908 |
Framework versions
- Transformers 4.33.2
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.13.3