electra_base_finetuned_ai4privacy_50k

This model is a fine-tuned version of google/electra-base-generator on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.3508
Overall Precision: 0.6649
Overall Recall: 0.6917
Overall F1: 0.6780
Overall Accuracy: 0.9118
Accountname F1: 0.8544
Accountnumber F1: 0.9850
Age F1: 0.8104
Amount F1: 0.3476
Bic F1: 0.6504
Bitcoinaddress F1: 0.8787
Buildingnumber F1: 0.4394
City F1: 0.7396
Companyname F1: 0.8249
Company Name F1: 0.0
County F1: 0.2505
Creditcardcvv F1: 0.0
Creditcardissuer F1: 0.0282
Creditcardnumber F1: 0.6742
Currency F1: 0.4509
Currencycode F1: 0.0
Currencyname F1: 0.0
Currencysymbol F1: 0.0101
Date F1: 0.7845
Dob F1: 0.0
Email F1: 0.9924
Ethereumaddress F1: 0.9923
Eyecolor F1: 0.0
Firstname F1: 0.7420
Fullname F1: 0.0
Gender F1: 0.4611
Height F1: 0.5351
Iban F1: 0.9890
Ip F1: 0.0
Ipv4 F1: 0.8270
Ipv6 F1: 0.8162
Jobarea F1: 0.1614
Jobdescriptor F1: 0.0
Jobtitle F1: 0.5561
Jobtype F1: 0.0071
Lastname F1: 0.4119
Litecoinaddress F1: 0.3699
Mac F1: 1.0
Maskednumber F1: 0.0034
Middlename F1: 0.0
Nearbygpscoordinate F1: 0.9987
Ordinaldirection F1: 0.0
Password F1: 0.9657
Phoneimei F1: 0.9923
Phonenumber F1: 0.8980
Phone Number F1: 0.0
Pin F1: 0.0
Prefix F1: 0.6811
Secondaryaddress F1: 0.9048
Sex F1: 0.7825
Ssn F1: 0.9841
State F1: 0.4807
Street F1: 0.7196
Streetaddress F1: 0.0
Suffix F1: 0.0
Time F1: 0.9357
Url F1: 0.9947
Useragent F1: 0.9760
Username F1: 0.9788
Vehiclevin F1: 0.9916
Vehiclevrm F1: 0.9249
Zipcode F1: 0.5165

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: cosine_with_restarts
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	Creditcardissuer F1	Creditcardnumber F1	Currency F1	Currencysymbol F1	Date F1	Email F1	Ethereumaddress F1	Firstname F1	Gender F1	Height F1	Iban F1	Ipv4 F1	Ipv6 F1	Jobarea F1	Jobtitle F1	Jobtype F1	Lastname F1	Litecoinaddress F1	Mac F1	Maskednumber F1	Nearbygpscoordinate F1	Password F1	Phoneimei F1	Phonenumber 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
No log	1.0	364	2.7775	0.2154	0.0923	0.1292	0.6859	0.0	0.0	0.0	0.0	0.0	0.3987	0.0	0.0	0.0	0.0	0.0	0.2333	0.0	0.0	0.0	0.0	0.5946	0.0	0.0	0.0	0.2597	0.0	0.5150	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.1433	0.0	0.1909	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.5481	0.0	0.0	0.0	0.0
3.3938	2.0	728	1.2022	0.4323	0.3798	0.4044	0.8191	0.0	0.6510	0.0	0.0655	0.0	0.8011	0.0138	0.0127	0.0	0.0	0.0	0.6484	0.0	0.0	0.5897	0.8698	0.9808	0.3339	0.0	0.0	0.9157	0.7386	0.7579	0.0	0.0	0.0	0.0	0.0	0.9026	0.0	0.9321	0.6538	0.9686	0.7228	0.0	0.0033	0.0	0.0045	0.0	0.0	0.0	0.9580	0.9094	0.3636	0.8841	0.0	0.2482
1.2855	3.0	1092	0.6727	0.5407	0.5144	0.5272	0.8586	0.0	0.9208	0.0793	0.1777	0.0	0.8201	0.2074	0.3516	0.6218	0.1078	0.0	0.1766	0.0090	0.0	0.7730	0.9847	0.9897	0.5398	0.0910	0.0	0.9707	0.8207	0.8089	0.0	0.1533	0.0	0.0	0.0	0.9854	0.0	0.9888	0.8804	0.9891	0.8733	0.0	0.5911	0.4069	0.5898	0.4945	0.0064	0.8508	0.9894	0.9590	0.8922	0.9793	0.6765	0.6717
1.2855	4.0	1456	0.4669	0.5617	0.6047	0.5824	0.8937	0.7388	0.9664	0.5581	0.3208	0.0608	0.9045	0.1664	0.5627	0.7413	0.1902	0.0	0.2043	0.3644	0.0	0.7788	0.9886	0.9747	0.6153	0.4966	0.2185	0.9430	0.8238	0.8162	0.0	0.4819	0.0	0.1742	0.6072	0.9941	0.0	0.9975	0.9214	0.9880	0.8925	0.0	0.6957	0.8821	0.9615	0.4664	0.1754	0.9038	0.9932	0.9573	0.8238	0.9665	0.6506	0.4045
0.6155	5.0	1820	0.3849	0.6189	0.6561	0.6370	0.9061	0.8456	0.9867	0.7757	0.3189	0.6517	0.9014	0.3436	0.6482	0.7471	0.2419	0.0088	0.5961	0.4163	0.0051	0.7790	0.9902	0.9923	0.6686	0.5717	0.3236	0.9846	0.8257	0.8162	0.0239	0.5419	0.0	0.1581	0.5098	1.0	0.0	0.9962	0.9568	0.9923	0.8970	0.1636	0.8670	0.8455	0.9773	0.4838	0.5196	0.9207	0.9939	0.9683	0.9219	0.9812	0.8868	0.5266
0.4334	6.0	2184	0.3554	0.6592	0.6864	0.6725	0.9105	0.8612	0.9833	0.7877	0.3533	0.5913	0.8747	0.4266	0.7182	0.8148	0.2510	0.0247	0.6709	0.4571	0.0051	0.7842	0.9924	0.9923	0.7345	0.4866	0.5521	0.9868	0.8270	0.8162	0.1288	0.5493	0.0071	0.3859	0.3726	1.0	0.0033	0.9987	0.9467	0.9923	0.8988	0.6138	0.8982	0.7983	0.9841	0.4826	0.6739	0.9342	0.9947	0.9769	0.9766	0.9916	0.928	0.5086
0.3827	7.0	2548	0.3508	0.6649	0.6917	0.6780	0.9118	0.8544	0.9850	0.8104	0.3476	0.6504	0.8787	0.4394	0.7396	0.8249	0.2505	0.0282	0.6742	0.4509	0.0101	0.7845	0.9924	0.9923	0.7420	0.4611	0.5351	0.9890	0.8270	0.8162	0.1614	0.5561	0.0071	0.4119	0.3699	1.0	0.0034	0.9987	0.9657	0.9923	0.8980	0.6811	0.9048	0.7825	0.9841	0.4807	0.7196	0.9357	0.9947	0.9760	0.9788	0.9916	0.9249	0.5165

Framework versions

Transformers 4.34.0
Pytorch 2.0.1+cu118
Datasets 2.14.5
Tokenizers 0.14.0

electra_base_finetuned_ai4privacy_50k

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

NSDT 3DConvert

UnrealSynth

DreamTexture.js