<!-- 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. -->
electra_large_finetuned_ai4privacy_50k
This model is a fine-tuned version of google/electra-large-generator on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1673
- Overall Precision: 0.8062
- Overall Recall: 0.8641
- Overall F1: 0.8342
- Overall Accuracy: 0.9461
- Accountname F1: 0.9864
- Accountnumber F1: 0.9982
- Age F1: 0.9268
- Amount F1: 0.9841
- Bic F1: 0.0314
- Bitcoinaddress F1: 0.9752
- Buildingnumber F1: 0.7392
- City F1: 0.9394
- Companyname F1: 0.8810
- Company Name F1: 0.0
- County F1: 0.9242
- Creditcardcvv F1: 0.4108
- Creditcardissuer F1: 0.9666
- Creditcardnumber F1: 0.3822
- Currency F1: 0.7147
- Currencycode F1: 0.0063
- Currencyname F1: 0.0
- Currencysymbol F1: 0.4777
- Date F1: 0.7894
- Dob F1: 0.0
- Email F1: 0.9995
- Ethereumaddress F1: 1.0
- Eyecolor F1: 0.9524
- Firstname F1: 0.9397
- Fullname F1: 0.0
- Gender F1: 0.9534
- Height F1: 0.9971
- Iban F1: 0.9894
- Ip F1: 0.0
- Ipv4 F1: 0.8392
- Ipv6 F1: 0.8234
- Jobarea F1: 0.8310
- Jobdescriptor F1: 0.0
- Jobtitle F1: 0.9091
- Jobtype F1: 0.9018
- Lastname F1: 0.9022
- Litecoinaddress F1: 0.9367
- Mac F1: 1.0
- Maskednumber F1: 0.0017
- Middlename F1: 0.6042
- Nearbygpscoordinate F1: 1.0
- Ordinaldirection F1: 0.9681
- Password F1: 0.9864
- Phoneimei F1: 0.9989
- Phonenumber F1: 0.8930
- Phone Number F1: 0.0
- Pin F1: 0.0790
- Prefix F1: 0.9276
- Secondaryaddress F1: 0.9991
- Sex F1: 0.9691
- Ssn F1: 0.9969
- State F1: 0.7368
- Street F1: 0.8879
- Streetaddress F1: 0.0
- Suffix F1: 0.0
- Time F1: 0.9913
- Url F1: 1.0
- Useragent F1: 0.9983
- Username F1: 0.9915
- Vehiclevin F1: 1.0
- Vehiclevrm F1: 0.7590
- Zipcode F1: 0.9740
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
- num_epochs: 5
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 |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
1.4042 | 1.0 | 1453 | 0.7245 | 0.3741 | 0.4495 | 0.4084 | 0.8546 | 0.0 | 0.6287 | 0.2233 | 0.1148 | 0.0 | 0.6852 | 0.0828 | 0.0081 | 0.0396 | 0.0 | 0.0023 | 0.0 | 0.0 | 0.5377 | 0.0 | 0.0 | 0.0 | 0.0045 | 0.7703 | 0.0 | 0.9762 | 0.7539 | 0.0 | 0.3610 | 0.0 | 0.0 | 0.0 | 0.7777 | 0.0 | 0.7198 | 0.7788 | 0.0 | 0.0 | 0.2118 | 0.0 | 0.0396 | 0.0 | 0.9599 | 0.0 | 0.0 | 0.9963 | 0.0 | 0.5189 | 0.9906 | 0.7997 | 0.0 | 0.0 | 0.0 | 0.7016 | 0.0067 | 0.4411 | 0.3806 | 0.0463 | 0.0 | 0.0 | 0.8291 | 0.9905 | 0.9454 | 0.0181 | 0.1640 | 0.0607 | 0.4242 |
0.4717 | 2.0 | 2906 | 0.3115 | 0.6916 | 0.7516 | 0.7204 | 0.9264 | 0.7970 | 0.9578 | 0.7357 | 0.8408 | 0.0124 | 0.8454 | 0.6145 | 0.5851 | 0.8553 | 0.0 | 0.1742 | 0.0 | 0.5833 | 0.6887 | 0.3724 | 0.0 | 0.0 | 0.4491 | 0.7878 | 0.0 | 0.9908 | 0.9940 | 0.0214 | 0.8468 | 0.0 | 0.0209 | 0.0057 | 0.9604 | 0.0 | 0.8372 | 0.8166 | 0.3298 | 0.0 | 0.7009 | 0.7820 | 0.8036 | 0.0 | 1.0 | 0.0040 | 0.0 | 1.0 | 0.7212 | 0.8235 | 0.9968 | 0.8803 | 0.0 | 0.0 | 0.9041 | 0.9774 | 0.6922 | 0.9644 | 0.5306 | 0.7743 | 0.0 | 0.0 | 0.9572 | 0.9929 | 0.9615 | 0.9672 | 0.6202 | 0.7051 | 0.8564 |
0.262 | 3.0 | 4359 | 0.2083 | 0.8049 | 0.8393 | 0.8218 | 0.9364 | 0.9577 | 0.9903 | 0.8979 | 0.9645 | 0.0 | 0.9601 | 0.6700 | 0.8740 | 0.8757 | 0.0 | 0.8088 | 0.0530 | 0.9163 | 0.6926 | 0.6291 | 0.0089 | 0.0 | 0.4640 | 0.7905 | 0.0 | 0.9979 | 1.0 | 0.7492 | 0.8940 | 0.0 | 0.8720 | 0.9371 | 0.9779 | 0.0 | 0.8397 | 0.8194 | 0.7801 | 0.0 | 0.8368 | 0.8449 | 0.8298 | 0.8777 | 1.0 | 0.0 | 0.0160 | 1.0 | 0.9680 | 0.9452 | 1.0 | 0.8901 | 0.0 | 0.0 | 0.9248 | 0.9973 | 0.9518 | 0.9889 | 0.7293 | 0.8828 | 0.0 | 0.0 | 0.9835 | 0.9984 | 0.9703 | 0.9892 | 0.9844 | 0.7865 | 0.9130 |
0.1949 | 4.0 | 5812 | 0.1728 | 0.8267 | 0.8641 | 0.8450 | 0.9408 | 0.9864 | 0.9982 | 0.9177 | 0.9756 | 0.0126 | 0.9735 | 0.7103 | 0.9387 | 0.8781 | 0.0 | 0.9146 | 0.1667 | 0.9678 | 0.6832 | 0.7022 | 0.0065 | 0.0 | 0.4788 | 0.7920 | 0.0 | 0.9995 | 1.0 | 0.9352 | 0.9272 | 0.0 | 0.9347 | 0.9971 | 0.9872 | 0.0 | 0.8397 | 0.8205 | 0.8226 | 0.0 | 0.9018 | 0.8937 | 0.8969 | 0.9447 | 1.0 | 0.0039 | 0.4836 | 1.0 | 0.9718 | 0.9794 | 0.9989 | 0.8938 | 0.0 | 0.0501 | 0.9197 | 0.9982 | 0.9675 | 0.9949 | 0.7357 | 0.8889 | 0.0 | 0.0 | 0.9913 | 0.9984 | 0.9983 | 0.9915 | 0.9974 | 0.7655 | 0.9681 |
0.1862 | 5.0 | 7265 | 0.1673 | 0.8062 | 0.8641 | 0.8342 | 0.9461 | 0.9864 | 0.9982 | 0.9268 | 0.9841 | 0.0314 | 0.9752 | 0.7392 | 0.9394 | 0.8810 | 0.0 | 0.9242 | 0.4108 | 0.9666 | 0.3822 | 0.7147 | 0.0063 | 0.0 | 0.4777 | 0.7894 | 0.0 | 0.9995 | 1.0 | 0.9524 | 0.9397 | 0.0 | 0.9534 | 0.9971 | 0.9894 | 0.0 | 0.8392 | 0.8234 | 0.8310 | 0.0 | 0.9091 | 0.9018 | 0.9022 | 0.9367 | 1.0 | 0.0017 | 0.6042 | 1.0 | 0.9681 | 0.9864 | 0.9989 | 0.8930 | 0.0 | 0.0790 | 0.9276 | 0.9991 | 0.9691 | 0.9969 | 0.7368 | 0.8879 | 0.0 | 0.0 | 0.9913 | 1.0 | 0.9983 | 0.9915 | 1.0 | 0.7590 | 0.9740 |
Framework versions
- Transformers 4.34.0
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.14.1